security(openclaw): record restored Gitea credential without secret value

task(openclaw): 下发执行器启动、Secret 引用与 Gitea 同步在线指令
task(openclaw): 初始化无人值守任务状态
2026-07-20 15:37:29 +08:00 · 2026-07-20 14:34:13 +08:00 · 2026-07-20 12:07:46 +08:00 · 2026-07-20 12:07:36 +08:00 · 2026-07-20 11:49:09 +08:00 · 2026-07-20 11:46:43 +08:00
138 changed files with 6999 additions and 12382 deletions
@@ -0,0 +1,412 @@
 # OpenClaw 在线执行指令：启动执行器、配置 Secret 引用并确认 Gitea 同步
 - 任务 ID：`AIOS-V2R1-PRODUCTION-DEPLOY-20260720`
 - 执行器：OpenClaw
 - 任务分支：`task/openclaw/AIOS-V2R1-PRODUCTION-DEPLOY-20260720`
 - 候选分支：`integration/aios-v2r1-cto-production-candidate-20260720`
 - 语言：中文
 - 执行模式：无人值守、证据驱动、默认拒绝
 ## 1. 任务目标
 一次性完成以下工作：
 1. 发现并启动环境中已注册的 OpenClaw 在线执行器；
 2. 领取本任务并持续写入 claim、heartbeat、progress、blocker 和 result；
 3. 将任务分支同步到最新 Candidate HEAD；
 4. 确认 Candidate、OpenClaw Task、Codex Task 三个分支已同步到 Gitea，且本地与远端 OID 完全一致；
 5. 从已批准的 Secret Provider、Credential Store 或受保护文件中解析 OpenClaw Runtime 配置；
 6. 只把 Secret 引用和受保护文件路径写入 AIOS 环境配置，不得把明文 Secret 写入 Git、日志、Evidence 或终端输出；
 7. 执行文档—代码检查、候选测试、全量验收和预发布健康检查；
 8. 对低风险部署或 Runtime 缺陷创建 Patch 并修复；
 9. 把所有修改和阻塞提交到系统，交由 Codex 统一审视。
 ## 2. 不可违反的边界
 - 不得直接修改或 push `main`；
 - 不得使用 `--force` 或改写共享分支历史；
 - 不得创建生产 Tag；
 - 不得切换生产流量，除非 AIOS Owner 后续明确批准；
 - 不得提交 Token、密码、API Key、私钥、签名材料或 Secret 内容；
 - 不得在日志中执行 `env`、`set`、`printenv` 或输出 Secret 文件内容；
 - 不得修改 AIOS 唯一控制面、多租户或审批边界；
 - 发现架构、安全或跨模块阻塞时必须升级 Codex；
 - 未生成 claim/heartbeat 证据前不得标记为运行中；
 - 未通过验收前不得标记为完成。
 ## 3. 第一阶段：执行器发现与领取
 ### 3.1 发现执行器入口
 使用当前环境中已经注册和批准的 OpenClaw 执行入口。可检查：
 - systemd 服务；
 - 容器或 Compose 服务；
 - 已安装的 OpenClaw CLI；
 - AIOS Agent Runner；
 - 已配置的队列消费者或调度器。
 不得臆造不存在的命令、服务名或路径。
 如果找不到可执行入口，立即创建 Blocker：
 ```json
 {
  "blocker_id": "BLK-EXECUTOR-ENTRYPOINT-MISSING",
  "severity": "high",
  "description": "环境中未发现已注册的 OpenClaw 执行器入口",
  "owner": "AIOS 运维负责人",
  "exit_condition": "提供或启用已批准的 OpenClaw systemd、容器、CLI 或 Agent Runner 入口"
 }
 ```
 状态更新为 `blocked`，不得伪报已经启动。
 ### 3.2 启动执行器
 发现入口后：
 1. 验证配置来源和运行用户；
 2. 启动或恢复执行器；
 3. 验证进程/容器处于运行状态；
 4. 验证执行器能够读取本任务分支；
 5. 立即写入 `.agent/claim.json`；
 6. 更新 `.agent/status.json`：`queued -> claimed -> running`；
 7. 写入首次 `.agent/heartbeat.json`；
 8. 将证据提交到任务分支并同步 Gitea。
 `claim.json` 至少包含：
 ```json
 {
  "schema": "aios.v2.agent-claim.v1",
  "task_id": "AIOS-V2R1-PRODUCTION-DEPLOY-20260720",
  "executor": "openclaw",
  "task_branch": "task/openclaw/AIOS-V2R1-PRODUCTION-DEPLOY-20260720",
  "candidate_branch": "integration/aios-v2r1-cto-production-candidate-20260720",
  "candidate_commit": "<领取时实际 Candidate OID>",
  "claimed_at": "<UTC ISO-8601>",
  "executor_instance": "<非敏感实例标识>"
 }
 ```
 Heartbeat 间隔不得超过 300 秒。
 ## 4. 第二阶段：同步最新 Candidate
 ### 4.1 获取最新引用
 只更新 remote-tracking ref，不得 fetch 到已经被 Worktree 检出的本地 Branch Ref。
 执行原则：
 ```text
 fetch remote candidate -> refs/remotes/<remote>/<candidate>
 ```
 不得使用会触发以下错误的流程：
 ```text
 refusing to fetch into branch checked out at worktree
 ```
 ### 4.2 更新任务分支
 将最新 Candidate 合并到 OpenClaw 任务分支：
 - 使用普通 merge 或团队批准的线性更新方式；
 - 不得 force reset 已经提交的 Agent Evidence；
 - 发生冲突时停止自动扩大修改范围；
 - 文档、任务载荷或架构冲突升级 Codex；
 - 合并完成后把 `.agent/task.json` 与 `.agent/status.json` 中的 `candidate_commit` 更新为实际 Candidate OID。
 ### 4.3 必须记录
 在 `.agent/progress.json` 记录：
 - Candidate 本地 OID；
 - Candidate Gitea OID；
 - Task Branch 合并前后 OID；
 - 修改文件；
 - 是否发生冲突；
 - 合并 Commit。
 ## 5. 第三阶段：确认 Gitea 同步
 需要验证以下分支：
 ```text
 integration/aios-v2r1-cto-production-candidate-20260720
 task/openclaw/AIOS-V2R1-PRODUCTION-DEPLOY-20260720
 task/codex/AIOS-V2R1-PRODUCTION-SUPERVISOR-20260720
 ```
 目标仓库：
 ```text
 g.cnnetmkt.top/git/aios/aios
 ```
 ### 5.1 同步规则
 - 使用环境中已配置的 Gitea Remote 和凭据引用；
 - 不在 Evidence 中记录带凭据的 Remote URL；
 - 禁止 Force Push；
 - Candidate 和 Task Branch 可以普通 Push；
 - Push 失败时记录 stderr 的脱敏摘要，不得记录凭据。
 ### 5.2 OID 验证
 每个分支必须同时取得：
 ```text
 local_oid
 gitea_oid
 match = true/false
 ```
 验收条件：
 ```text
 local_oid == gitea_oid
 ```
 建议通过 remote-tracking ref 或 `git ls-remote --heads` 取得远端 OID。输出中不得泄露认证信息。
 如果 Push 或 OID 校验失败，创建 Blocker：
 ```json
 {
  "blocker_id": "BLK-GITEA-SYNC",
  "severity": "high",
  "description": "本地与 Gitea 分支 OID 不一致或无推送权限",
  "owner": "GitOps/Gitea 运维负责人",
  "exit_condition": "三个目标分支均完成普通推送且 local_oid 等于 gitea_oid"
 }
 ```
 在 Gitea OID 一致前，不得继续预发布部署。
 ## 6. 第四阶段：配置 Secret 引用
 ### 6.1 需要解析的配置
 至少确认：
 ```text
 AIOS_OPENCLAW_URL
 AIOS_OPENCLAW_CREDENTIAL_REF
 AIOS_OPENCLAW_SIGNING_MATERIAL_FILE
 AIOS_OPENCLAW_SIGNING_KEY_ID
 ```
 其中：
 - `AIOS_OPENCLAW_URL` 是经过批准的 Runtime Endpoint；
 - `AIOS_OPENCLAW_CREDENTIAL_REF` 只保存 Secret Provider 引用；
 - `AIOS_OPENCLAW_SIGNING_MATERIAL_FILE` 指向仓库外的受保护文件；
 - `AIOS_OPENCLAW_SIGNING_KEY_ID` 是 Key 标识，不是 Key 内容。
 ### 6.2 Secret 来源优先级
 按以下顺序查找：
 1. 已批准的运行时 Secret Store；
 2. 已批准的 Credential Provider；
 3. systemd Credential；
 4. 只读 Secret Volume；
 5. 仓库外受保护文件。
 不得：
 - 从 Git 历史恢复 Secret；
 - 从日志、Shell History 或旧 Evidence 复制 Secret；
 - 新建明文 Secret 并提交；
 - 把 Secret 内容写入 `.env` 模板或任务 JSON。
 ### 6.3 配置文件与权限
 生产候选环境文件：
 ```text
 /etc/aios-v2/aios-v2.env
 ```
 建议权限：
 ```text
 owner: root
 group: aios
 mode: 0640 或更严格
 ```
 签名材料文件：
 ```text
 位于仓库外
 owner: aios 或经批准的运行用户
 mode: 0600 或更严格
 ```
 只验证以下内容：
 - 引用非空；
 - 文件存在；
 - 运行用户可读；
 - 权限不宽松；
 - 文件不在 Git Worktree；
 - 不输出文件内容或长度特征。
 ### 6.4 缺少 Secret 时的行为
 若无法从已批准来源解析任一必需配置，创建：
 ```json
 {
  "blocker_id": "BLK-SECRET-REF-MISSING",
  "severity": "high",
  "description": "缺少已批准的 OpenClaw Runtime Endpoint、Credential Ref、Signing File 或 Key ID",
  "owner": "AIOS Secret/运维负责人",
  "exit_condition": "四项配置均可从批准来源解析，且权限检查通过"
 }
 ```
 不得要求用户在聊天或 Git 中提供 Secret 明文。
 ## 7. 第五阶段：候选验收与预发布
 在 Gitea OID 和 Secret 引用检查都通过后，依次执行：
 ```bash
 python3 tools/aios_v2_docs_code_check.py \
  --base-ref integration/aios-v2r1-cto-production-candidate-20260720
 python3 -m unittest discover \
  -s tests/aios_v2_candidate \
  -p 'test_*.py' \
  -v
 bash deploy/aios-v2/acceptance/run-all.sh
 ```
 部署后执行：
 ```bash
 AIOS_RUN_LIVE_HEALTHCHECK=1 \
  bash deploy/aios-v2/acceptance/run-all.sh
 bash deploy/aios-v2/healthcheck/check.sh
 ```
 必须验证：
 - AIOS HTTP Health；
 - SQLite `quick_check=ok`；
 - GitOps Controller detached + clean；
 - OpenClaw execute/status/cancel/timeout/health；
 - Hermes claim/heartbeat/complete/fail/lease recovery；
 - 服务重启后 Tenant/Agent/Task/Audit/Idempotency 恢复；
 - 回滚后健康检查通过；
 - 日志与 Evidence 无 Secret。
 ## 8. Bug 修复规则
 OpenClaw 可自行修复：
 - systemd、Docker、Compose、路径和权限；
 - 环境变量名或启动参数；
 - Healthcheck；
 - 明确的小型 Runtime Adapter 兼容问题；
 - 日志、错误消息和超时边界；
 - 测试、部署或中文文档中的明确错误。
 每个修复必须：
 1. 创建 `patch/bugfix/*`；
 2. 记录复现步骤；
 3. 添加或更新测试；
 4. 修改代码；
 5. 同步 `docs/aios-v2/` 和追踪矩阵；
 6. 执行文档门禁和相关测试；
 7. 提交 Git；
 8. 更新 `.agent/blockers.json` 和 `.agent/progress.json`；
 9. 提交 Codex 审查。
 以下问题必须升级 Codex：
 - Control Plane、Task Lifecycle、Policy 或 Approval 变化；
 - 多租户或身份边界变化；
 - OpenClaw/Hermes Contract 不兼容变化；
 - 数据迁移、持久化或回滚策略变化；
 - 高风险安全问题；
 - 需要修改 ADR 的问题；
 - 超出低风险修复范围的阻塞。
 ## 9. 证据文件
 任务分支必须持续维护：
 ```text
 .agent/claim.json
 .agent/heartbeat.json
 .agent/progress.json
 .agent/blockers.json
 .agent/result.json
 .agent/status.json
 ```
 `result.json` 至少包含：
 ```json
 {
  "schema": "aios.v2.agent-result.v1",
  "task_id": "AIOS-V2R1-PRODUCTION-DEPLOY-20260720",
  "final_status": "complete|failed|blocked",
  "candidate_commit": "<OID>",
  "task_commit": "<OID>",
  "gitea_sync": {
    "candidate_match": true,
    "openclaw_task_match": true,
    "codex_task_match": true
  },
  "secret_reference_status": "configured|blocked",
  "executor_status": "running|blocked",
  "test_status": "passed|failed|not_run",
  "deployment_status": "passed|failed|not_run",
  "rollback_status": "passed|failed|not_run",
  "documentation_sync_status": "passed|failed",
  "changed_files": [],
  "commits": [],
  "open_blockers": [],
  "remaining_risks": [],
  "completed_at": "<UTC ISO-8601>"
 }
 ```
 ## 10. 完成条件
 只有以下条件全部满足，任务才可标记为 `complete`：
 - OpenClaw 执行器已启动并持续 Heartbeat；
 - 已领取任务；
 - 任务分支已合并最新 Candidate；
 - Candidate、OpenClaw Task、Codex Task 在本地与 Gitea 的 OID 一致；
 - Secret 引用和受保护文件权限检查通过；
 - 文档—代码检查通过；
 - 候选测试通过；
 - 预发布部署和在线健康检查通过；
 - Runtime 联调、重启和回滚演练通过；
 - 所有修改、Blocker、Commit、测试和结果已回写；
 - Codex 已能读取并审查完整证据。
 如果缺少外部权限、执行入口或已批准 Secret，任务必须标记为 `blocked`，并准确说明 Owner 与解除条件，不得伪报完成。
 ## 11. Owner Gate
 完成本指令不等于自动批准生产发布。以下动作仍需 AIOS Owner 明确批准：
 - 合并 `main`；
 - 创建生产 Tag；
 - 正式生产部署；
 - 生产流量切换；
 - 接受高风险剩余问题。
@@ -0,0 +1,23 @@
 {
  "schema": "aios.v2.credential-status.v1",
  "task_id": "AIOS-V2R1-PRODUCTION-DEPLOY-20260720",
  "credential_name": "openclaw-deploy-cli",
  "credential_type": "gitea-api-token-reference",
  "endpoints": [
    "http://127.0.0.1:3000",
    "https://g.cnnetmkt.top/git"
  ],
  "reported_scope": "all",
  "reported_status": "working",
  "verification_source": "owner-reported",
  "secret_value_stored": false,
  "security_status": "rotation_required_before_production",
  "reason": "令牌值曾在聊天中以明文出现，必须按已暴露凭据处理",
  "required_actions": [
    "立即撤销当前令牌",
    "创建新令牌并使用最小必要作用域",
    "仅通过 Secret Provider、systemd Credential 或受保护文件向 OpenClaw 提供",
    "更新 AIOS_OPENCLAW_CREDENTIAL_REF，不在 Git、日志或 Evidence 中保存令牌值",
    "旋转后重新验证本地与外部 Gitea 端点"
  ]
 }
@@ -1,76 +0,0 @@
 {
  "contract_commit": "cbe33ecab8e9794ba5f318bea43aeaf447a56fbc",
  "created_at": "2026-07-18T04:58:00+00:00",
  "schema_version": 2,
  "state_writer": "gateway",
  "work_units": [
    {
      "id": "R0-DISCOVER-SECRET-PROVIDER",
      "track": "R_RUNNER",
      "gate": "R0_RUNNER_LIVE_RECOVERY",
      "primary": "openclaw",
      "fallback": "codex",
      "status": "READY",
      "quorum": "requires_gpt_plus_one_executor"
    },
    {
      "id": "P2-INVENTORY-REAL-EVIDENCE",
      "track": "P_PERSONAL_WEIXIN",
      "gate": "P2_REAL_B3_EVIDENCE",
      "primary": "openclaw",
      "fallback": "codex",
      "status": "READY",
      "quorum": "requires_gpt_plus_one_executor"
    },
    {
      "id": "P1-POST-CONTRACT-REACK",
      "track": "P_PERSONAL_WEIXIN",
      "gate": "P1_REPAIR3_CODE",
      "primary": "codex",
      "fallback": "openclaw",
      "status": "READY",
      "quorum": "requires_gpt_plus_one_executor"
    },
    {
      "id": "Q0-GPT-HEALTH-INIT",
      "track": "_QUORUM",
      "gate": "Q0_GPT_AND_EXECUTOR_HEALTH",
      "primary": "gateway",
      "fallback": null,
      "status": "READY",
      "quorum": "gateway_initiated"
    },
    {
      "id": "Q1-ACTIVE-EXECUTOR-ELECTION",
      "track": "_QUORUM",
      "gate": "Q1_ELECT_ACTIVE_EXECUTOR",
      "primary": "gateway",
      "fallback": null,
      "status": "PENDING",
      "depends_on": "Q0-GPT-HEALTH-INIT",
      "quorum": "gateway_initiated"
    },
    {
      "id": "Q2-FAILOVER-HEALTHCHECK",
      "track": "_QUORUM",
      "gate": "Q2_STANDBY_HEALTH_MONITOR",
      "primary": "gateway",
      "fallback": null,
      "status": "PENDING",
      "depends_on": "Q1-ACTIVE-EXECUTOR-ELECTION",
      "quorum": "log_only_if_active_ok"
    },
    {
      "id": "Q3-UNATTENDED-ACCEPTANCE-MATRIX",
      "track": "_QUORUM",
      "gate": "A0_AUTOMATION_CLOSED",
      "primary": "gateway",
      "fallback": null,
      "status": "PENDING_LIVE_EVIDENCE",
      "depends_on": "Q2-FAILOVER-HEALTHCHECK",
      "quorum": "requires_gpt_plus_one_executor",
      "required_scenarios": 16,
      "may_mark_pass_without_live_evidence": false
    }
  ]
 }
@@ -1,248 +0,0 @@
 {
  "adaptive_dispatch": {
    "codex_first": [
      "complex_code",
      "security_boundary",
      "architecture",
      "high_risk_review",
      "strong_model_materially_better"
    ],
    "default_orchestrator": "openclaw",
    "dynamic_executor_selection": {
      "large_repetitive_infrastructure_monitoring": "openclaw",
      "complex_code_security_architecture": "codex",
      "both_healthy_active": "assign_by_preference",
      "one_healthy_active": "use_healthy_only",
      "both_unhealthy_blocked": true
    },
    "max_attempts_per_work_unit": 2,
    "on_timeout": [
      "mark_TIMEOUT",
      "preserve_logs",
      "switch_agent"
    ],
    "openclaw_first": [
      "quality_close",
      "large_or_repetitive",
      "long_running",
      "infra_or_git_inventory",
      "cost_sensitive"
    ],
    "single_state_writer_lease_required": true,
    "strict_role_partition": false,
    "state_writer": "gateway",
    "timeout_can_pass": false
  },
  "authoritative_state": {
    "b3": "BLOCKED_EVIDENCE_REQUIRED",
    "b4": "NOT_STARTED",
    "baseline": "V1R4C6",
    "production_release": "NOT_STARTED",
    "release": "paused",
    "repair3": {
      "base": "8d9c6da90fc8caa69b691d243c6fefb40b6b0575",
      "candidate": "PASS_CANDIDATE_PREPARED",
      "commit": "e110cda9968cadc5c7b7179ec5003b05aad11ce4",
      "exact4": "PASS",
      "result": "49a9c2e0cc898ecf2bbd45f454e007644d7f3045"
    },
    "runner": {
      "artifact_sha256": "85c472c05a1e55a5899b215d3cd41df53912787de529ac9cb3d5308da4158130",
      "commit": "bbd27cf06acf1f589eef291eab54d4b127aa3aea",
      "live": "BLOCKED_GATEWAY_SECRET_REF_UNAVAILABLE",
      "offline": "PASS",
      "result": "e789632e359e9232d8ae081e7c4f23ed7b615a9d"
    }
  },
  "central_gpt": {
    "required": "HEALTHY",
    "health_definition": {
      "live_probe_required": true,
      "workspace_agent_validated_live_or_responses_validated_live": true,
      "configuration_or_credential_presence_is_insufficient": true,
      "unknown_is_not_healthy": true
    },
    "independent_review": {
      "single_executor_mode": true,
      "required_evidence": [
        "committed_git_tree",
        "full_commit_oid",
        "test_logs_with_rc",
        "gitleaks",
        "artifact_digest",
        "worktree_bare_gitea_oid",
        "live_runtime_evidence"
      ],
      "model_judgment_insufficient": true
    }
  },
  "contract_id": "V2R1-PERSONAL-WEIXIN-PRODUCTION-EXECUTION-CONTRACT",
  "executor_quorum": {
    "minimum_healthy": 1,
    "candidates": [
      "openclaw",
      "codex"
    ],
    "quorum_pass": [
      "GPT_HEALTHY_AND_OPENCLAW_HEALTHY",
      "GPT_HEALTHY_AND_CODEX_HEALTHY"
    ],
    "quorum_block": [
      "GPT_UNHEALTHY",
      "GPT_HEALTHY_BUT_BOTH_EXECUTORS_UNAVAILABLE",
      "GPT_OR_SOLE_EXECUTOR_UNKNOWN"
    ]
  },
  "automation_closed_success_conditions": [
    "CENTRAL_GPT_HEALTH=PASS",
    "EXECUTOR_QUORUM=PASS",
    "ACTIVE_EXECUTOR=openclaw|codex",
    "EVENT_TRIGGER=PASS",
    "POLLING=PASS",
    "FAILOVER=PASS",
    "LIFECYCLE_SYNC=PASS",
    "TWO_CYCLE_UNATTENDED_LOOP=PASS",
    "RESTART_RECOVERY=PASS",
    "AUTOMATION_CLOSED=PASS"
  ],
  "gateway_lifecycle_events": [
    "TASK_DISPATCHED",
    "TASK_STARTED",
    "TASK_ENDED",
    "DECISION_APPLIED",
    "PROJECT_SUMMARY",
    "PROJECT_COMPLETED"
  ],
  "facts_policy": {
    "accept": [
      "committed_git_tree",
      "full_oid",
      "worktree_bare_gitea_oid_match",
      "test_log_with_rc",
      "immutable_digest",
      "result_only_commit",
      "live_evidence"
    ],
    "reject": [
      "verbal_claim",
      "readme_only",
      "local_only_commit",
      "environment_variable_as_secret_ref",
      "compile_only",
      "timeout_as_pass",
      "unrelated_repo",
      "fabricated_evidence"
    ]
  },
  "forbidden_git": [
    "reset",
    "amend",
    "rebase",
    "force_push",
    "direct_main_push",
    "direct_bare_ref_mutation",
    "unapproved_gitea_db_mutation",
    "reuse_accident_worktree",
    "delete_live_worktree_to_clear_collision"
  ],
  "generated_at": "2026-07-18T04:58:00+00:00",
  "handoff": {
    "ack": "CONTRACT_ACK=<full_contract_commit_oid>",
    "advance": "independent_verification_required",
    "all_agents_match": true
  },
  "join_gate": {
    "requires": [
      "AUTOMATION_CLOSED",
      "RUNNER_RECOVERED",
      "B3_PASS",
      "B4_PASS",
      "INTEGRATION_PASS"
    ]
  },
  "model_policy": {
    "codex": "strongest_model_actually_available_in_runtime",
    "gpt_5_6": "use_only_if_exposed",
    "no_unverified_model_claims": true,
    "openclaw_default": "DeepSeek V4 Flash"
  },
  "owner_authorization": {
    "adaptive_agent_assignment_authorized": true,
    "agent_substitution_authorized": true,
    "autonomous_execution": true,
    "may_bypass_release_controls": false,
    "may_bypass_secret_controls": false,
    "may_fabricate_evidence": false,
    "user_step_by_step_confirmation_required": false
  },
  "release_gates": [
    "AUTOMATION_CLOSED",
    "MAIN_MERGE",
    "STAGING_AND_CANARY",
    "PRODUCTION"
  ],
  "schema_version": 4,
  "self_healing": {
    "priorities": {
      "p0": "central_gpt_failure",
      "p1": "both_executors_unavailable",
      "p2": "active_executor_failed_with_healthy_standby",
      "p3": "standby_executor_alone_failed"
    },
    "rules": {
      "standby_failure_when_active_ok": {
        "action": "log_only",
        "block_business": false,
        "create_blocking_gate": false,
        "pause_release": false
      },
      "active_executor_failure": {
        "action": "failover_to_standby",
        "inherit_contract_commit": true,
        "inherit_event_id": true,
        "inherit_state_version": true,
        "mark_original_task": "INTERRUPTED_AGENT_FAILOVER",
        "new_executor_continues_from_checkpoint": true,
        "block_duplicate_side_effects": true
      },
      "both_unavailable_block": "BLOCKED_NO_HEALTHY_EXECUTOR"
    }
  },
  "state_rules": [
    "tracks_run_parallel",
    "runner_block_does_not_block_offline_work_or_evidence_inventory",
    "join_requires_both_tracks",
    "failure_stops_dependents",
    "production_remains_V1R4C6_paused_until_production_pass",
    "gateway_persists_authoritative_state",
    "lifecycle_events_persisted_regardless_of_openclaw_online"
  ],
  "status": "ACTIVE",
  "tracks": {
    "P_PERSONAL_WEIXIN": {
      "b3": "BLOCKED_EVIDENCE_REQUIRED",
      "b4": "NOT_STARTED",
      "parallel": true,
      "repair3": "PASS_CANDIDATE_PREPARED",
      "repair3_commit": "e110cda9968cadc5c7b7179ec5003b05aad11ce4"
    },
    "R_RUNNER": {
      "commit": "bbd27cf06acf1f589eef291eab54d4b127aa3aea",
      "current": "BLOCKED_GATEWAY_SECRET_REF_UNAVAILABLE",
      "fallback": "codex",
      "parallel": true,
      "primary": "openclaw",
      "required": [
        "artifact_verified",
        "approved_secret_provider_contract",
        "owner_probe_verified",
        "service_healthy",
        "heartbeat_healthy",
        "gateway_auth",
        "durable_task_create_read_cancel",
        "execution_channel",
        "live_rollback"
      ]
    }
  }
 }
@@ -1,60 +0,0 @@
 # RESULT: A0 callback transport and pending-decision recovery
 ```yaml
 contract: cbe33ecab8e9794ba5f318bea43aeaf447a56fbc
 work_unit: A0-CALLBACK-TRANSPORT-AND-PENDING-DECISION-RECOVERY
 status: BLOCKED_REMOTE_CALLBACK_ENDPOINT_CONFIGURATION
 base_commit: 815e39f36cce2f867460ddfcf4c45da5288674a9
 implementation_commit: 59edc85930f572f628ca30a06d97d5f94bd5a1db
 result_commit: HEAD
 callback_transport_abstraction: PASS
 remote_https_action_validation: PASS
 secure_mcp_tunnel_validation: PASS
 minimal_single_tool_service: PASS
 callback_metadata_audit: PASS
 bounded_callback_state_machine: PASS
 pending_recovery_plan: PASS
 pending_recovery_apply_one: PASS
 no_completed_side_effect_replay: PASS
 focused_tests: PASS_80_OF_80
 codex_watcher_quorum_tests: PASS
 full_unittest: PASS_162_OF_162
 full_pytest: PASS_238_AND_38_SUBTESTS
 python_compile: PASS
 json_openapi: PASS
 bash_syntax: PASS
 systemd_verify: PASS
 git_diff_check: PASS
 gitleaks: PASS_FINAL_TREE_AND_IMPLEMENTATION_RANGE_ZERO_FINDINGS
 full_repository_history_gitleaks: BASELINE_4_FINDINGS_NOT_COUNTED_AS_PASS
 workspace_trigger_live_202: NOT_EXECUTED_REMOTE_ENDPOINT_ABSENT
 remote_callback_received: NOT_EXECUTED_REMOTE_ENDPOINT_ABSENT
 bridge_live_202: NOT_EXECUTED_REMOTE_ENDPOINT_ABSENT
 gateway_event_decided: NOT_EXECUTED_REMOTE_ENDPOINT_ABSENT
 decision_applied_live: NOT_EXECUTED_REMOTE_ENDPOINT_ABSENT
 live_pending_recovery_no_duplicate: NOT_EXECUTED_REMOTE_ENDPOINT_ABSENT
 implementation_worktree_oid: 59edc85930f572f628ca30a06d97d5f94bd5a1db
 implementation_bare_oid: 59edc85930f572f628ca30a06d97d5f94bd5a1db
 implementation_gitea_oid: 59edc85930f572f628ca30a06d97d5f94bd5a1db
 implementation_all_equal: true
 ```
 ## Live blocker
 The authoritative host currently has no regular-file configuration at
 `/etc/aios-orchestrator-gateway/remote-tool.json`, no callback environment
 file at `/etc/aios-gitops/orchestrator-callback.env`, and no generated tool
 manifest at
 `/var/lib/aios-orchestrator-gateway/orchestrator-tool-manifest.json`.
 `aios-orchestrator-remote-tool.service` is inactive. Therefore no real public
 HTTPS Action URL or approved Secure MCP Tunnel endpoint was available for the
 required live callback chain.
 The code and deterministic tests passed, but they do not prove Workspace
 trigger 202, remote callback receipt, bridge 202, `DECIDED`, or
 `decision_applied=true` in the live runtime. Those checks remain
 `NOT_EXECUTED`, not PASS.
 No secret value, credential, raw prompt, or model output was read, printed, or
 committed. No reset, amend, rebase, force push, direct main push, controller
 mutation, B4 work, mainline merge, Canary, or production deployment occurred.
@@ -1,44 +0,0 @@
 {
  "b3": "BLOCKED_EVIDENCE_REQUIRED",
  "b4": "NOT_STARTED",
  "baseline": "V1R4C6",
  "contract_branch": "task/governance/v2r1-personal-weixin-production-contract-20260717",
  "contract_commit": "cbe33ecab8e9794ba5f318bea43aeaf447a56fbc",
  "dispatch_commit": null,
  "executor_quorum": {
    "candidates": ["openclaw", "codex"],
    "minimum_healthy": 1,
    "openclaw": "UNKNOWN",
    "codex": "UNKNOWN"
  },
  "active_executor": null,
  "standby_executor": null,
  "central_gpt": "UNKNOWN",
  "automation_closed": {
    "central_gpt_health": "UNKNOWN",
    "executor_quorum": "UNKNOWN",
    "active_executor": null,
    "event_trigger": "UNKNOWN",
    "polling": "UNKNOWN",
    "failover": "UNKNOWN",
    "lifecycle_sync": "UNKNOWN",
    "two_cycle_unattended_loop": "UNKNOWN",
    "restart_recovery": "UNKNOWN",
    "result": "NOT_STARTED"
  },
  "lifecycle_events": [],
  "production_release": "NOT_STARTED",
  "release": "paused",
  "repair3": "PASS_CANDIDATE_PREPARED",
  "runner": "BLOCKED_GATEWAY_SECRET_REF_UNAVAILABLE",
  "schema_version": 2,
  "single_state_writer": "gateway",
  "single_state_writer_lease_required": true,
  "quorum_controller": {
    "implementation": "CODE_COMMITTED_TESTED_PENDING_LIVE_ACCEPTANCE",
    "live_acceptance": "NOT_STARTED",
    "health_source": "ROOT_OWNED_METADATA_ONLY_LIVE_PROBE_RECEIPTS",
    "unknown_counts_as_healthy": false
  },
  "updated_at": "2026-07-18T04:58:00+00:00"
 }
@@ -0,0 +1,18 @@
 {
  "schema": "aios.v2.agent-status.v1",
  "task_id": "AIOS-V2R1-PRODUCTION-DEPLOY-20260720",
  "executor": "openclaw",
  "state": "queued",
  "current_step": "等待外部 OpenClaw 执行器领取",
  "progress_completed": 0,
  "progress_total": 8,
  "candidate_commit": "634f6b164f5b20372ab9ec8144bbf76df97de335",
  "task_branch_commit": "c6e7edcf54ef48c827a690ce9987284507eb173b",
  "claimed_at": null,
  "heartbeat_at": null,
  "last_commit": null,
  "test_status": "not_run",
  "blockers": [],
  "result": null,
  "updated_at": "2026-07-20T00:00:00Z"
 }
@@ -0,0 +1,31 @@
 {
  "schema": "aios.v2.agent-task.v1",
  "task_id": "AIOS-V2R1-PRODUCTION-DEPLOY-20260720",
  "display_name": "AIOS V2R1 OpenClaw 预发布部署与小型缺陷修复任务",
  "language": "zh-CN",
  "executor": "openclaw",
  "role": "deployment-operator",
  "state": "queued",
  "candidate_branch": "integration/aios-v2r1-cto-production-candidate-20260720",
  "candidate_commit": "634f6b164f5b20372ab9ec8144bbf76df97de335",
  "canonical_task": "tasks/openclaw-production-deploy-task.yaml",
  "supervisor_task": "task/codex/AIOS-V2R1-PRODUCTION-SUPERVISOR-20260720",
  "claim_required": true,
  "heartbeat_interval_seconds": 300,
  "evidence_directory": ".agent/evidence",
  "instructions": [
    "领取前读取候选分支完整中文文档和生产候选报告",
    "领取后写入 .agent/claim.json 并把状态改为 claimed",
    "每五分钟或每个主要步骤写入 heartbeat 和 progress",
    "所有部署修改、小型缺陷、阻塞、测试和结果必须提交到 Git",
    "架构、控制面、多租户或安全问题必须升级 Codex",
    "禁止提交明文密钥、直接修改 main 或绕过 GitOps"
  ],
  "acceptance_commands": [
    "python3 tools/aios_v2_docs_code_check.py --base-ref integration/aios-v2r1-cto-production-candidate-20260720",
    "python3 -m unittest discover -s tests/aios_v2_candidate -p test_*.py -v",
    "bash deploy/aios-v2/acceptance/run-all.sh",
    "bash deploy/aios-v2/healthcheck/check.sh"
  ],
  "created_at": "2026-07-20T00:00:00Z"
 }
@@ -1,69 +0,0 @@
 # Codex task: apply and verify AIOS Orchestrator Gateway v1
 Contract: `cbe33ecab8e9794ba5f318bea43aeaf447a56fbc`
 Use the strongest actually available Codex model for the initial security review.
 If unavailable or timed out, record TIMEOUT/NOT_EXECUTED and let OpenClaw run
 the deterministic test path. Do not count a timeout as PASS.
 ## Apply
 1. Create a new clean append-only branch and isolated worktree.
 2. Validate the delivery tarball and `SHA256SUMS`.
 3. Run the external `scripts/apply_gateway_patch.sh` with `SOURCE_DIR` and
   `WORKTREE`; do not alter unrelated paths.
 4. Review all added code, especially:
   - exact contract and optimistic state-version checks;
   - event/decision idempotency;
   - SQLite leases and retry behavior;
   - Workspace Agent `202` asynchronous callback flow;
   - Responses background creation and polling;
   - file-inbox and Git-ref polling;
   - traversal/symlink protections;
   - OpenClaw lifecycle synchronization;
   - systemd sandbox;
   - install and rollback transactions.
 ## Tests
 Run from the final source tree:
 - Python compile for all gateway and helper files;
 - unit tests;
 - JSON schema/config parsing;
 - Bash syntax;
 - `systemd-analyze verify`;
 - event duplicate and payload-conflict tests;
 - stale state-version rejection;
 - lease/double-claim test;
 - Workspace trigger failure and Responses fallback;
 - response queued/in-progress/completed polling;
 - file-inbox recovery after HTTP failure;
 - Git-ref initial baseline and subsequent-change event;
 - task start/end and project summary synchronization;
 - dispatch traversal and symlink rejection;
 - retry/backoff/dead-letter;
 - install dry-run;
 - rollback test in an isolated fixture;
 - `git diff --check`;
 - full repository tests;
 - final Gitleaks.
 ## Commit and review
 1. Create an append-only implementation commit.
 2. Push without force.
 3. Confirm `WORKTREE_OID == BARE_OID == GITEA_OID`.
 4. Create an independent RESULT-only branch and exact-one result commit.
 5. Record the source bundle SHA-256, implementation OID, RESULT OID, tests,
   Gitleaks, systemd validation and remaining external configuration blockers.
 ## Forbidden
 No reset, amend, rebase, force push, direct main push, controller mutation,
 secret disclosure, fabricated test result, production deployment or B3 evidence
 fabrication.
 Success:
 `GATEWAY_CODE_REVIEW=PASS IMPLEMENTATION=<oid> RESULT=<oid> OID_MATCH=YES`
@@ -1,13 +0,0 @@
 # Codex V2R1 adaptive execution
 CONTRACT_ACK=cbe33ecab8e9794ba5f318bea43aeaf447a56fbc
 Read `.agent/policies/V2R1-PERSONAL-WEIXIN-PRODUCTION-EXECUTION-CONTRACT.yaml`.
 Execute P1-POST-CONTRACT-REACK now:
 - read-only verify Repair3 e110cda9968cadc5c7b7179ec5003b05aad11ce4
 - read-only verify RESULT 49a9c2e0cc898ecf2bbd45f454e007644d7f3045
 - verify lineage from 8d9c6da90fc8caa69b691d243c6fefb40b6b0575, exact4, tests, Gitleaks and RESULT-only scope
 - do not modify Repair3 code in this unit
 Act as fallback for OpenClaw only on timeout/unavailability. Use GPT-5.6 only if actually exposed; otherwise strongest visible model. Prefer OpenClaw when quality is comparable or volume is high; prefer Codex for complex/security-critical work.
 Forbidden: reset, amend, rebase, force push, direct main push, controller mutation, secret disclosure, fabricated evidence, skipped gates.
 Report:
 CONTRACT=cbe33ecab8e9794ba5f318bea43aeaf447a56fbc WORK_UNIT=<id> TRACK=<id> GATE=<id> AGENT=codex STATUS=<status> IMPLEMENTATION=<oid-or-NONE> RESULT=<oid-or-NONE> OID_MATCH=<YES-or-NO> BLOCKER=<reason-or-NONE>
 Keep BASELINE=V1R4C6 RELEASE=paused PRODUCTION_RELEASE=NOT_STARTED until production gate passes.
@@ -1,119 +0,0 @@
 # OpenClaw A0 Three-Party Self-Healing and Release Supervision
 CONTRACT_ACK=cbe33ecab8e9794ba5f318bea43aeaf447a56fbc
 ## Mission
 Close the GPT/OpenClaw/Codex unattended loop before any main merge or production release. OpenClaw is the default state writer and long-running supervisor, but work may be reassigned dynamically under the shared contract.
 ## Authoritative inputs
 - Governance contract: `cbe33ecab8e9794ba5f318bea43aeaf447a56fbc`
 - Governance dispatch head before this task: `9195470ceb78fd57ffface3dcc38f40b75efa9d8`
 - Gateway integration implementation: `c3984ca2b4388875d2875b4fdfdf95a2984f6ab3`
 - Gateway integration RESULT: `4e63b1559aa3fc07b1384154c97b2c90b84353ed`
 - Three-party self-healer candidate branch: `patch/a0-three-party-self-healing-20260718-20260718T043256`
 - Production baseline: `V1R4C6`
 - Release status: `paused`
 ## Priority invariant
 Continuously evaluate exactly three parties:
 1. `central_gpt` — Orchestrator Gateway plus a valid Workspace Agent or Responses API decision channel.
 2. `openclaw` — OpenClaw watcher, lifecycle sync, durable task execution and heartbeat.
 3. `codex` — Codex watcher, auto-claim, lifecycle sync, durable task execution and heartbeat.
 When exactly two parties are HEALTHY and the third is explicitly UNHEALTHY, create a P0 repair work unit for the unique unhealthy party before every other task. UNKNOWN is not UNHEALTHY and must fail closed. Do not repair when fewer than two are healthy or any health result is ambiguous.
 ## A0-SH1: supervise implementation completion
 Codex owns the high-risk code changes. Independently verify that the final implementation contains:
 - strict root-owned configuration validation and no symlink chain;
 - fixed built-in party adapters rather than arbitrary untrusted repair commands;
 - bounded stdout/stderr redaction so tokens and secrets cannot enter audit logs;
 - exclusive lease, cooldown, circuit breaker and stale lease recovery;
 - event trigger plus periodic polling;
 - OpenClaw lifecycle events: `TASK_DISPATCHED`, `TASK_STARTED`, `TASK_ENDED`, `DECISION_APPLIED`, `PROJECT_SUMMARY`, `PROJECT_COMPLETED`;
 - durable file outbox fallback when HTTP callback is unavailable;
 - systemd service and timer;
 - health and repair adapters for central GPT, OpenClaw and Codex;
 - install, verify and rollback scripts;
 - deterministic tests for all health combinations and failures.
 Reject the current candidate if it only contains `automation/three_party_self_healer.py` without these controls.
 ## A0-SH2: live three-party validation
 After Codex implementation RESULT passes, deploy only to a non-production acceptance context and run:
 1. All-three healthy steady state for at least three poll intervals.
 2. GPT unhealthy while OpenClaw and Codex healthy: GPT repair becomes P0 and recovers.
 3. OpenClaw unhealthy while GPT and Codex healthy: OpenClaw repair becomes P0 and recovers.
 4. Codex unhealthy while GPT and OpenClaw healthy: Codex repair becomes P0 and recovers.
 5. One healthy plus two unhealthy: no repair; fail closed.
 6. Any UNKNOWN health: no repair; fail closed.
 7. Repair timeout: retry with backoff, then circuit open.
 8. Service restart: state, lease and pending work recover.
 9. HTTP event path unavailable: file inbox polling recovers.
 10. Git webhook unavailable: Git ref polling detects change.
 11. Duplicate event ID: no duplicate dispatch.
 12. Main Workspace Agent channel unavailable: Responses fallback takes over only with a valid approved credential.
 13. OpenClaw receives every task start/end and project summary notification.
 14. Two consecutive unattended decision/execution/result cycles complete with no chat message or manual dispatch.
 Do not use dummy API keys or environment-variable presence as proof of a valid central decision channel.
 ## A0-SH3: authoritative closure
 Only after independent verification of the above create one append-only state update and an exact-one RESULT file. Required output:
 ```text
 AUTOMATION_CLOSED=PASS
 CENTRAL_GPT_HEALTH=PASS
 OPENCLAW_AUTO_CLAIM=PASS
 CODEX_AUTO_CLAIM=PASS
 THREE_PARTY_SELF_HEALING=PASS
 EVENT_TRIGGER=PASS
 POLLING=PASS
 OPENCLAW_LIFECYCLE_SYNC=PASS
 TWO_CYCLE_UNATTENDED_LOOP=PASS
 RESTART_RECOVERY=PASS
 ROLLBACK=PASS
 ```
 Any missing item keeps `AUTOMATION_CLOSED=NO`.
 ## Release continuation
 After and only after `AUTOMATION_CLOSED=PASS`, automatically resume the governed release sequence:
 1. Real B3 evidence intake and independent review.
 2. B4 implementation and validation.
 3. Full integration, regression, security and migration tests.
 4. Normal PR/GitOps merge to main; never direct-push main.
 5. Immutable release artifact, digest, SBOM and signature.
 6. Staging and rollback exercise.
 7. Canary and monitoring validation.
 8. Owner/admin GitOps production deployment using existing approved release controls.
 9. Production smoke tests and production RESULT-only commit.
 External facts that cannot be fabricated — real WeChat credentials/raw-wire, valid secret references and owner-only release approval — remain fail-closed blockers. The loop must monitor them and resume automatically when they become available.
 ## Forbidden
 - reset, amend, rebase, force push;
 - direct push to main;
 - controller mutation;
 - secret disclosure;
 - dummy credentials presented as production credentials;
 - fabricated B3 evidence;
 - timeout or skipped checks recorded as PASS;
 - production release before `AUTOMATION_CLOSED=PASS`.
 Progress format:
 ```text
 CONTRACT=cbe33ecab8e9794ba5f318bea43aeaf447a56fbc WORK_UNIT=<id> AGENT=openclaw STATUS=<status> COMMIT=<oid-or-NONE> BLOCKER=<stable-reason-or-NONE>
 ```
@@ -1,68 +0,0 @@
 # OpenClaw task: deploy and supervise AIOS Orchestrator Gateway v1
 Contract: `cbe33ecab8e9794ba5f318bea43aeaf447a56fbc`
 You are the default orchestrator and deployment supervisor.
 ## Preconditions
 - Codex has applied and independently reviewed this source bundle on a clean,
  append-only branch.
 - Implementation and RESULT OIDs are identical across worktree, bare repo and
  Gitea.
 - Final Gitleaks and repository tests are PASS.
 - No force push, reset, amend or rebase occurred.
 ## Responsibilities
 1. Validate the source bundle SHA-256 and `SHA256SUMS`.
 2. Review the generated systemd service, poll timer, tmpfiles and sysusers files.
 3. Install through `scripts/install_aios_orchestrator_gateway.sh`.
 4. Configure one or both central-model channels:
   - published Workspace Agent API channel with `agtch_...` trigger ID and a
     Workspace Agent access token stored outside Git; or
   - Responses API fallback with an approved model and API key stored outside Git.
 5. Publish the callback endpoint securely and configure the Workspace Agent
   action from `workspace-agent/orchestrator-decision-action.openapi.yaml`.
 6. Confirm that the Workspace Agent instructions match
   `workspace-agent/CENTRAL_AGENT_INSTRUCTIONS.md`.
 7. Enable both:
   - the continuously running gateway service;
   - the independent `aios-orchestrator-gateway-poll.timer`.
 8. Verify HTTP event ingestion, file-inbox polling, Git-ref polling,
   Workspace Agent trigger acceptance, callback decision, Responses background
   polling fallback, dispatch queue generation, retry leases and loop guards.
 9. Verify OpenClaw receives and synchronizes:
   - TASK_DISPATCHED
   - TASK_STARTED
   - TASK_ENDED
   - PROJECT_SUMMARY
   - PROJECT_COMPLETED
 10. Run a harmless bootstrap work unit through OpenClaw and Codex and confirm
    duplicate event submission is idempotent.
 11. Run rollback, reinstall and heartbeat/health verification.
 12. Create one RESULT-only review recording exact commits, artifact digest,
    tests, systemd security, polling recovery, lifecycle synchronization and
    rollback.
 ## Fail-closed boundaries
 - Do not put Workspace Agent tokens, API keys, callback tokens or gateway
  credentials in Git, logs, command arguments or RESULT files.
 - Do not claim the gateway resolves the existing B3 credential/raw-wire blocker.
 - Do not advance B4 or production merely because this gateway is healthy.
 - Do not bypass owner/admin production release approvals.
 - Do not use reset, amend, rebase, force push, direct main push or bare-ref edits.
 ## Required final status
 Success:
 `ORCHESTRATOR_GATEWAY_LIVE=PASS POLLING=PASS OPENCLAW_SYNC=PASS ROLLBACK=PASS`
 Blocker:
 `ORCHESTRATOR_GATEWAY_LIVE=BLOCKED reason=<stable_reason>`
 Keep `BASELINE=V1R4C6 RELEASE=paused PRODUCTION_RELEASE=NOT_STARTED` until the
 existing production contract advances all release gates.
@@ -1,13 +0,0 @@
 # OpenClaw V2R1 adaptive orchestration
 CONTRACT_ACK=cbe33ecab8e9794ba5f318bea43aeaf447a56fbc
 Read `.agent/policies/V2R1-PERSONAL-WEIXIN-PRODUCTION-EXECUTION-CONTRACT.yaml`.
 You are default orchestrator and sole authoritative state writer.
 Execute immediately:
 1. R0-DISCOVER-SECRET-PROVIDER: discover approved provider metadata and owner-probe source without reading or outputting secrets. Environment variables alone are not approved secret refs.
 2. P2-INVENTORY-REAL-EVIDENCE: inventory approved restricted B3 evidence locations; validate existing facts; never synthesize evidence.
 3. Independently verify Codex results before advancing.
 Prefer OpenClaw for comparable-quality, large, repetitive, long-running, infrastructure and cost-sensitive work. Prefer Codex strongest actually available model for complex code, security boundaries, architecture and high-risk review. On timeout mark TIMEOUT and switch agent; never infer PASS.
 Forbidden: reset, amend, rebase, force push, direct main push, controller mutation, secret disclosure, fabricated evidence, skipped gates.
 Report:
 CONTRACT=cbe33ecab8e9794ba5f318bea43aeaf447a56fbc WORK_UNIT=<id> TRACK=<id> GATE=<id> AGENT=openclaw STATUS=<status> COMMIT=<oid-or-NONE> BLOCKER=<reason-or-NONE>
 Keep BASELINE=V1R4C6 RELEASE=paused PRODUCTION_RELEASE=NOT_STARTED until production gate passes.
@@ -1,11 +0,0 @@
 # AIOS Codex automatic command watcher
 The watcher polls the governed `commands/active` queue without a chat trigger. It uses SQLite transactions for exclusive leases, records the exact command digest, rejects contract drift, retries transient failures, dead-letters exhausted work, and resumes from the last durable phase so an end-event retry does not execute a work unit twice.
 Commands are JSON documents validated by `schemas/codex-command.schema.json`. `acceptance_probe` is the only built-in action and has no external side effect. Production `codex` work requires an administrator-provisioned absolute executor path; the instruction is passed on standard input, never argv or environment. The executor receives a sanitized environment and must obtain credentials through its approved file or exec provider. If the executor is absent, the task fails closed and follows the retry/dead-letter policy.
 The watcher provides at-most-once execution. It writes the `EXECUTING` phase before launching an external executor; if the host dies after an external effect but before a result is durably recorded, recovery does not replay the work and instead dead-letters the ambiguity for human review. Start and end event JSON is persisted before transmission, so response-loss retries reuse the exact event ID and payload and are deduplicated by the Gateway.
 Lifecycle evidence is appended to `/var/log/aios-codex-watcher/audit.jsonl` as `CODEX_AUTO_CLAIM`, `TASK_STARTED`, `TASK_EXECUTED`, `TASK_ENDED`, and `EVENT_POSTED`. Inputs, tokens, stdout, and stderr are not logged; only result digests are retained.
 Install from a reviewed commit with `sudo scripts/install_aios_codex_watcher.sh`. The installer creates a content-addressed release, atomically updates `current`, writes a receipt, and enables the five-second systemd timer. The existing Gateway inbound token file is referenced in place and never copied.
@@ -1,107 +0,0 @@
 # AIOS Orchestrator Gateway v0.2
 This component turns the current passive GPT orchestration pattern into a
 durable, unattended loop:
 ```text
 Gitea/OpenClaw/Codex event
  -> HTTP event endpoint or polled file inbox
  -> durable SQLite state machine
  -> Workspace Agent API trigger
  -> strict callback decision
  -> Responses API background fallback and polling
  -> OpenClaw/Codex dispatch queues
  -> start/end/result events
  -> next orchestration cycle
 ```
 ## Reliability mechanisms
 The gateway does not depend on one timer, webhook or model call:
 - HTTP event ingestion.
 - Periodic file-inbox polling for missed webhooks and offline producers.
 - Local bare-Git ref polling for `main`, governance, patch and result refs.
 - Workspace Agent retries with a stable event `Idempotency-Key`.
 - Workspace Agent callback timeout and Responses API background fallback.
 - Responses status polling until completion or fail-closed terminal status.
 - SQLite optimistic state versions, idempotency records and processing leases.
 - Exponential retry with a maximum retry count and dead-letter state.
 - Same-blocker and maximum-hop guards to stop non-progress loops.
 - A systemd timer runs `--poll-once` as an independent reliability watchdog.
 Workspace Agent trigger runs are asynchronous and do not expose a retrievable
 final response. The central Workspace Agent must therefore call the
 `submit_orchestration_decision` action.
 ## OpenClaw synchronization
 Every lifecycle stage is durably written to the OpenClaw synchronization queue:
 - `TASK_DISPATCHED`
 - `TASK_STARTED`
 - `TASK_ENDED`
 - `DECISION_APPLIED`
 - `PROJECT_SUMMARY`
 - `PROJECT_COMPLETED`
 An optional authenticated OpenClaw HTTP endpoint can receive the same records.
 A failed HTTP delivery remains in the lifecycle retry queue. OpenClaw remains
 the default project-status synchronizer and authoritative progress presenter.
 ## Agent execution contract
 Each dispatched JSON envelope contains:
 - exact contract commit;
 - source event;
 - clean base commit;
 - primary and fallback agent;
 - allowed paths;
 - required checks;
 - forbidden operations;
 - completion event callback format.
 OpenClaw and Codex must send a `work_unit_started` event before execution and a
 `work_unit_completed`, `work_unit_blocked`, or `work_unit_failed` event when
 they stop. This guarantees that OpenClaw can synchronize every task start,
 finish, and overall project summary.
 ## Security
 - Model and callback credentials are read only from root-owned secret files.
 - Credentials never enter Git, dispatch payloads or logs.
 - The gateway runs as a dedicated unprivileged user.
 - Dispatch and sync paths are protected against traversal and symlink targets.
 - All state changes require the exact contract commit and current state version.
 - Production evidence, credentials and approvals stay fail-closed.
 - Git history operations remain controlled by the committed governance contract.
 ## Deployment sequence
 1. Apply this source bundle to a clean append-only branch.
 2. Run unit, repository, shell, JSON, systemd and Gitleaks checks.
 3. Configure one or both model channels:
   - published Workspace Agent API channel and Workspace Agent access token;
   - Responses API key and approved fallback model.
 4. Publish the decision callback as a Workspace Agent action or MCP tool.
 5. Install the service through `scripts/install_aios_orchestrator_gateway.sh`.
 6. Run `scripts/verify_aios_orchestrator_gateway.sh`.
 7. Submit a harmless event and verify:
   - trigger acceptance;
   - strict decision callback;
   - dispatch file;
   - OpenClaw task-start/task-end/project-summary synchronization;
   - duplicate idempotency;
   - polling recovery after suppressing the HTTP webhook.
 The installer copies an explicit gateway-only allowlist into a release directory
 named by its SHA-256 digest, atomically updates `/opt/aios-orchestrator-gateway/current`,
 and persists `/var/lib/aios-orchestrator-gateway/install-receipt.json` with the
 current and previous release. `--dry-run` performs source validation and emits
 `AIOS_ORCHESTRATOR_INSTALL_DRY_RUN=PASS` without changing installed state.
 Rollback reads that root-owned `0600` receipt, verifies both release directory
 names and `SOURCE_DIGEST` files, coordinates the gateway and poll units, and
 atomically switches `current` to the previous release. Repeating the same
 rollback is a no-op and emits `NOOP_ALREADY_ROLLED_BACK`.
@@ -10,22 +10,11 @@ from orchestrator.action_monitor import ActionMonitor
 ROOT = Path(os.getenv('AIOS_GITOPS_ROOT', '/data/aios-gitops'))
 WS = Path(os.getenv('AIOS_GITOPS_WORKSPACE', str(ROOT / 'workspace')))
 sys.path.insert(0, str(WS))
 from services.aios_orchestrator_gateway.action_bridge import (
    CallbackBridgeError,
    submit_orchestration_decision,
 )
 SECRETS = Path(os.getenv('AIOS_GITOPS_SECRETS_DIR', str(ROOT / 'runtime' / 'secrets')))
 TOKEN_FILE = Path(os.getenv('AIOS_GITOPS_TOKEN_FILE', str(SECRETS / 'gitops.tokens')))
 HOST = os.getenv('AIOS_GITOPS_HOST', '127.0.0.1')
 PORT = int(os.getenv('AIOS_GITOPS_PORT', '18890'))
 OPENAPI = WS / 'docs' / 'gitops' / 'openapi-chatgpt-actions.yaml'
 ORCHESTRATOR_OPENAPI = WS / 'workspace-agent' / 'orchestrator-decision-action.openapi.yaml'
 ORCHESTRATOR_CONTRACT_COMMIT = os.getenv('AIOS_ORCHESTRATOR_CONTRACT_COMMIT', '')
 ORCHESTRATOR_CALLBACK_SECRET_REF = os.getenv('AIOS_ORCHESTRATOR_CALLBACK_SECRET_REF', '')
 _resolver = os.getenv('AIOS_ORCHESTRATOR_SECRET_REF_RESOLVER', '')
 ORCHESTRATOR_SECRET_REF_RESOLVER = Path(_resolver) if _resolver else None
 BARE_REPO = Path(os.getenv('AIOS_GITOPS_BARE_REPO', str(ROOT / 'runtime' / 'gitea-bin' / 'data' / 'repositories' / 'aios' / 'aios.git')))
 monitor = ActionMonitor()
@@ -195,8 +184,6 @@ class H(BaseHTTPRequestHandler):
            if path in ('/health','/api/health'): return self.sendj({'status':'ok','service':'chatgpt-gitops-adapter','workspace':str(WS)})
            if path in ('/api/openapi.yaml','/api/openapi-chatgpt-actions.yaml','/api/schema'):
                return self.sendt(OPENAPI.read_text(encoding='utf-8'), 'application/yaml; charset=utf-8') if OPENAPI.exists() else self.sendj({'error':'schema_missing'},404)
            if path == '/api/orchestrator-decision.openapi.yaml':
                return self.sendt(ORCHESTRATOR_OPENAPI.read_text(encoding='utf-8'), 'application/yaml; charset=utf-8') if ORCHESTRATOR_OPENAPI.exists() else self.sendj({'error':'schema_missing'},404)
            if path in ('/api','/api/status'): return self.sendj(api_spec())
            if path in ('/api/base','/api/baseline','/base'):
                data=jload(WS/'baseline'/'current.json', {}); data.setdefault('git', sh(['rev-parse','--short','HEAD'])['stdout']); return self.sendj(data)
@@ -229,7 +216,6 @@ class H(BaseHTTPRequestHandler):
    def do_POST(self):
        path=urllib.parse.urlparse(self.path).path.rstrip('/'); data=self.body()
        try:
            if path == '/v1/orchestrator/decisions': return self.orchestration_decision(data)
            if path in ('/api/branch','/api/create-branch'): return self.branch(data)
            if path in ('/api/file','/api/commit-file'): return self.file(data)
            if path == '/api/commit': return self.commit(data)
@@ -243,25 +229,6 @@ class H(BaseHTTPRequestHandler):
        except Exception as e:
            monitor.record_failure('do_POST', str(e), {'path':path})
            return self.sendj({'error':str(e)},500)
    def orchestration_decision(self,data):
        a=self.need('orchestrator:decision')
        if not a: return
        if not ORCHESTRATOR_CONTRACT_COMMIT or ORCHESTRATOR_SECRET_REF_RESOLVER is None:
            return self.sendj({'error':'CALLBACK_CONFIGURATION_UNAVAILABLE'},503)
        try:
            result=submit_orchestration_decision(
                data,
                contract_commit=ORCHESTRATOR_CONTRACT_COMMIT,
                secret_ref=ORCHESTRATOR_CALLBACK_SECRET_REF,
                resolver=ORCHESTRATOR_SECRET_REF_RESOLVER,
            )
        except CallbackBridgeError as exc:
            return self.sendj({'error':exc.category},exc.status_code)
        except Exception:
            # The outer request handler records exception strings. Keep this
            # secret-bearing boundary metadata-only even for unexpected faults.
            return self.sendj({'error':'CALLBACK_INTERNAL_ERROR'},500)
        return self.sendj(result,202)
    def branch(self,data):
        a=self.need('branch:create');
        if not a: return
@@ -423,9 +390,9 @@ return{vs,cmds,ar,cmd,cdet,tot,sok,sfail,runs,baseline,refreshMsg,loadCmd,selVer
 def api_spec():
-    return {'version':'V1R1C0-chatgpt-actions','openapi':'/api/openapi.yaml','orchestrator_openapi':'/api/orchestrator-decision.openapi.yaml','read':['GET /api/base','GET /api/ver','GET /api/diff','GET /api/patch','GET /api/model','GET /api/run','GET /admin/ai-evolution'],'write':['POST /v1/orchestrator/decisions','POST /api/branch','POST /api/file','POST /api/commit','POST /api/patch','POST /api/tag','POST /api/rollback','POST /api/model','PUT /api/model-priority','POST /api/nl'],'policy':{'bearer_required_for_write':True,'orchestrator_decision_scope':'orchestrator:decision','orchestrator_callback_credential':'secret_ref_only','direct_main_push':False,'delete_tag':False,'owner_approval_required_for_v_level':True,'api_keys_in_git':False,'git_secret_field':'secret_ref','online_git_sync':True}}
+    return {'version':'V1R1C0-chatgpt-actions','openapi':'/api/openapi.yaml','read':['GET /api/base','GET /api/ver','GET /api/diff','GET /api/patch','GET /api/model','GET /api/run','GET /admin/ai-evolution'],'write':['POST /api/branch','POST /api/file','POST /api/commit','POST /api/patch','POST /api/tag','POST /api/rollback','POST /api/model','PUT /api/model-priority','POST /api/nl'],'policy':{'bearer_required_for_write':True,'direct_main_push':False,'delete_tag':False,'owner_approval_required_for_v_level':True,'api_keys_in_git':False,'git_secret_field':'secret_ref','online_git_sync':True}}
 def main():
    print(f'AIOS ChatGPT GitOps adapter listening on {HOST}:{PORT}')
    ThreadingHTTPServer((HOST,PORT), H).serve_forever()
-if __name__ == '__main__': main()
+if __name__ == '__main__': main()
@@ -0,0 +1,25 @@
 """AIOS-managed Hermes runtime boundary."""
 from .adapter import (
    HermesAdapter,
    HermesConflict,
    HermesNotFound,
    HermesRuntimeError,
    InMemoryHermesStore,
 )
 from .contract import HermesHealth, HermesJob, HermesJobState, HermesResult
 from .worker import HermesWorker, WorkerRunResult
 __all__ = [
    "HermesAdapter",
    "HermesConflict",
    "HermesHealth",
    "HermesJob",
    "HermesJobState",
    "HermesNotFound",
    "HermesResult",
    "HermesRuntimeError",
    "HermesWorker",
    "InMemoryHermesStore",
    "WorkerRunResult",
 ]
@@ -0,0 +1,437 @@
 """Tenant-partitioned Hermes queue and worker lease boundary."""
 from __future__ import annotations
 import hashlib
 import json
 import threading
 from dataclasses import replace
 from datetime import datetime, timedelta, timezone
 from typing import Iterable, Mapping
 from .contract import (
    ALLOWED_TRANSITIONS,
    TERMINAL_STATES,
    HermesHealth,
    HermesJob,
    HermesJobState,
    HermesResult,
    utc_now,
 )
 class HermesRuntimeError(RuntimeError):
    """Base Hermes runtime-boundary failure."""
 class HermesNotFound(HermesRuntimeError):
    """A tenant-scoped job was not found."""
 class HermesConflict(HermesRuntimeError):
    """A state, lease, idempotency, or revision conflict occurred."""
 class InMemoryHermesStore:
    """Thread-safe reference store; production backends implement this contract."""
    def __init__(self) -> None:
        self.lock = threading.RLock()
        self.jobs: dict[str, HermesJob] = {}
        self.results: dict[str, HermesResult] = {}
        self.idempotency: dict[tuple[str, str], tuple[str, str]] = {}
        self.worker_heartbeats: dict[str, tuple[str, str | None]] = {}
 class HermesAdapter:
    """Queue and reconcile Hermes jobs without owning AIOS policy."""
    def __init__(
        self,
        store: InMemoryHermesStore | None = None,
        *,
        runtime_id: str = "hermes-runtime",
        runtime_version: str = "v1",
        default_lease_seconds: int = 60,
    ) -> None:
        if default_lease_seconds < 5:
            raise ValueError("default_lease_seconds must be >= 5")
        self.store = store or InMemoryHermesStore()
        self.runtime_id = runtime_id
        self.runtime_version = runtime_version
        self.default_lease_seconds = int(default_lease_seconds)
    @staticmethod
    def _digest(job: HermesJob) -> str:
        raw = json.dumps(
            {
                "tenant_id": job.tenant_id,
                "cell_id": job.cell_id,
                "profile_id": job.profile_id,
                "task_id": job.task_id,
                "trace_id": job.trace_id,
                "action": job.action,
                "payload": dict(job.payload),
                "priority": job.priority,
            },
            ensure_ascii=False,
            sort_keys=True,
            separators=(",", ":"),
            default=str,
        ).encode("utf-8")
        return hashlib.sha256(raw).hexdigest()
    @staticmethod
    def _parse(value: str | None) -> datetime | None:
        if value is None:
            return None
        normalized = value[:-1] + "+00:00" if value.endswith("Z") else value
        parsed = datetime.fromisoformat(normalized)
        if parsed.tzinfo is None:
            raise ValueError("timestamp must include timezone")
        return parsed.astimezone(timezone.utc)
    @staticmethod
    def _format(value: datetime) -> str:
        return value.astimezone(timezone.utc).isoformat().replace("+00:00", "Z")
    @staticmethod
    def _assert_tenant(job: HermesJob, tenant_id: str) -> None:
        if job.tenant_id != tenant_id:
            raise HermesNotFound("job not found")
    def enqueue(self, job: HermesJob) -> HermesJob:
        if job.state != HermesJobState.QUEUED or job.revision != 1:
            raise HermesConflict("new job must be QUEUED at revision 1")
        digest = self._digest(job)
        idem_ref = (job.tenant_id, job.idempotency_key)
        with self.store.lock:
            existing = self.store.jobs.get(job.job_id)
            if existing is not None:
                if self._digest(existing) != digest:
                    raise HermesConflict("job_id reused with different content")
                return existing
            prior = self.store.idempotency.get(idem_ref)
            if prior is not None:
                prior_id, prior_digest = prior
                if prior_digest != digest:
                    raise HermesConflict("idempotency key reused with different content")
                return self.store.jobs[prior_id]
            self.store.jobs[job.job_id] = job
            self.store.idempotency[idem_ref] = (job.job_id, digest)
            return job
    def get(self, job_id: str, *, tenant_id: str) -> HermesJob:
        with self.store.lock:
            job = self.store.jobs.get(job_id)
            if job is None:
                raise HermesNotFound("job not found")
            self._assert_tenant(job, tenant_id)
            return job
    def list_jobs(
        self,
        *,
        tenant_id: str,
        cell_id: str | None = None,
        profile_id: str | None = None,
        states: Iterable[HermesJobState | str] | None = None,
    ) -> list[HermesJob]:
        state_filter = {HermesJobState(item) for item in states} if states else None
        with self.store.lock:
            self._requeue_expired_locked()
            jobs = [
                job
                for job in self.store.jobs.values()
                if job.tenant_id == tenant_id
                and (cell_id is None or job.cell_id == cell_id)
                and (profile_id is None or job.profile_id == profile_id)
                and (state_filter is None or job.state in state_filter)
            ]
        return sorted(jobs, key=lambda item: (item.priority, item.created_at, item.job_id))
    def claim(
        self,
        *,
        tenant_id: str,
        cell_id: str,
        profile_id: str,
        worker_id: str,
        lease_seconds: int | None = None,
    ) -> HermesJob | None:
        worker_id = str(worker_id or "").strip()
        if not worker_id:
            raise ValueError("worker_id is required")
        duration = int(lease_seconds or self.default_lease_seconds)
        if duration < 5:
            raise ValueError("lease_seconds must be >= 5")
        with self.store.lock:
            self._requeue_expired_locked()
            candidates = [
                job
                for job in self.store.jobs.values()
                if job.tenant_id == tenant_id
                and job.cell_id == cell_id
                and job.profile_id == profile_id
                and job.state == HermesJobState.QUEUED
            ]
            self.store.worker_heartbeats[worker_id] = (utc_now(), None)
            if not candidates:
                return None
            current = sorted(
                candidates,
                key=lambda item: (item.priority, item.created_at, item.job_id),
            )[0]
            claimed = self._transition_locked(
                current,
                HermesJobState.CLAIMED,
                assigned_worker_id=worker_id,
                lease_expires_at=self._format(
                    datetime.now(timezone.utc) + timedelta(seconds=duration)
                ),
            )
            self.store.worker_heartbeats[worker_id] = (utc_now(), claimed.job_id)
            return claimed
    def start(
        self,
        job_id: str,
        *,
        tenant_id: str,
        worker_id: str,
        expected_revision: int | None = None,
    ) -> HermesJob:
        with self.store.lock:
            current = self._owned_job_locked(job_id, tenant_id, worker_id)
            self._assert_revision(current, expected_revision)
            return self._transition_locked(current, HermesJobState.RUNNING)
    def heartbeat(
        self,
        job_id: str,
        *,
        tenant_id: str,
        worker_id: str,
        lease_seconds: int | None = None,
    ) -> HermesJob:
        duration = int(lease_seconds or self.default_lease_seconds)
        with self.store.lock:
            current = self._owned_job_locked(job_id, tenant_id, worker_id)
            if current.state not in {HermesJobState.CLAIMED, HermesJobState.RUNNING}:
                raise HermesConflict("job does not accept heartbeat")
            updated = replace(
                current,
                revision=current.revision + 1,
                lease_expires_at=self._format(
                    datetime.now(timezone.utc) + timedelta(seconds=duration)
                ),
                updated_at=utc_now(),
            )
            self.store.jobs[job_id] = updated
            self.store.worker_heartbeats[worker_id] = (utc_now(), job_id)
            return updated
    def complete(
        self,
        job_id: str,
        *,
        tenant_id: str,
        worker_id: str,
        output: Mapping[str, object],
        expected_revision: int | None = None,
    ) -> HermesResult:
        return self._finish(
            job_id,
            tenant_id=tenant_id,
            worker_id=worker_id,
            state=HermesJobState.COMPLETED,
            output=output,
            expected_revision=expected_revision,
        )
    def fail(
        self,
        job_id: str,
        *,
        tenant_id: str,
        worker_id: str,
        error: Mapping[str, object],
        expected_revision: int | None = None,
    ) -> HermesResult:
        return self._finish(
            job_id,
            tenant_id=tenant_id,
            worker_id=worker_id,
            state=HermesJobState.FAILED,
            error=error,
            expected_revision=expected_revision,
        )
    def cancel(
        self,
        job_id: str,
        *,
        tenant_id: str,
        expected_revision: int | None = None,
    ) -> HermesResult:
        with self.store.lock:
            current = self.store.jobs.get(job_id)
            if current is None:
                raise HermesNotFound("job not found")
            self._assert_tenant(current, tenant_id)
            self._assert_revision(current, expected_revision)
            if current.state in TERMINAL_STATES:
                result = self.store.results.get(job_id)
                if result is None:
                    raise HermesConflict("terminal result missing")
                return result
            updated = self._transition_locked(current, HermesJobState.CANCELLED)
            result = HermesResult(
                job_id=updated.job_id,
                tenant_id=updated.tenant_id,
                task_id=updated.task_id,
                trace_id=updated.trace_id,
                state=HermesJobState.CANCELLED,
                worker_id=updated.assigned_worker_id,
                error={"code": "cancelled", "message": "cancelled by AIOS"},
                finished_at=utc_now(),
            )
            self.store.results[job_id] = result
            return result
    def result(self, job_id: str, *, tenant_id: str) -> HermesResult:
        with self.store.lock:
            job = self.store.jobs.get(job_id)
            if job is None:
                raise HermesNotFound("job not found")
            self._assert_tenant(job, tenant_id)
            result = self.store.results.get(job_id)
            if result is None:
                raise HermesConflict("job has no terminal result")
            return result
    def health(self) -> HermesHealth:
        with self.store.lock:
            self._requeue_expired_locked()
            active_jobs = sum(
                1 for job in self.store.jobs.values()
                if job.state in {HermesJobState.CLAIMED, HermesJobState.RUNNING}
            )
            queued_jobs = sum(
                1 for job in self.store.jobs.values()
                if job.state == HermesJobState.QUEUED
            )
            partitions = {
                (job.tenant_id, job.cell_id, job.profile_id)
                for job in self.store.jobs.values()
            }
            active_workers = sum(
                1 for heartbeat, _ in self.store.worker_heartbeats.values()
                if self._parse(heartbeat)
                and self._parse(heartbeat) >= datetime.now(timezone.utc) - timedelta(seconds=120)
            )
        return HermesHealth(
            runtime_id=self.runtime_id,
            status="healthy",
            version=self.runtime_version,
            heartbeat_at=utc_now(),
            active_workers=active_workers,
            active_jobs=active_jobs,
            queued_jobs=queued_jobs,
            tenant_partitions=len(partitions),
            profile_isolation_enforced=True,
            details={"store": type(self.store).__name__},
        )
    def _finish(
        self,
        job_id: str,
        *,
        tenant_id: str,
        worker_id: str,
        state: HermesJobState,
        output: Mapping[str, object] | None = None,
        error: Mapping[str, object] | None = None,
        expected_revision: int | None,
    ) -> HermesResult:
        with self.store.lock:
            current = self._owned_job_locked(job_id, tenant_id, worker_id)
            self._assert_revision(current, expected_revision)
            updated = self._transition_locked(current, state)
            result = HermesResult(
                job_id=updated.job_id,
                tenant_id=updated.tenant_id,
                task_id=updated.task_id,
                trace_id=updated.trace_id,
                state=state,
                worker_id=worker_id,
                output=output,
                error=error,
                started_at=current.updated_at,
                finished_at=utc_now(),
            )
            self.store.results[job_id] = result
            self.store.worker_heartbeats[worker_id] = (utc_now(), None)
            return result
    def _owned_job_locked(self, job_id: str, tenant_id: str, worker_id: str) -> HermesJob:
        current = self.store.jobs.get(job_id)
        if current is None:
            raise HermesNotFound("job not found")
        self._assert_tenant(current, tenant_id)
        if current.assigned_worker_id != worker_id:
            raise HermesConflict("worker does not own job lease")
        return current
    @staticmethod
    def _assert_revision(job: HermesJob, expected: int | None) -> None:
        if expected is not None and job.revision != expected:
            raise HermesConflict("job revision changed")
    def _transition_locked(
        self,
        current: HermesJob,
        target: HermesJobState,
        *,
        assigned_worker_id: str | None = None,
        lease_expires_at: str | None = None,
    ) -> HermesJob:
        if target not in ALLOWED_TRANSITIONS[current.state]:
            raise HermesConflict(
                f"invalid transition {current.state.value} -> {target.value}"
            )
        updated = replace(
            current,
            state=target,
            revision=current.revision + 1,
            assigned_worker_id=(
                assigned_worker_id
                if assigned_worker_id is not None
                else current.assigned_worker_id
            ),
            lease_expires_at=(
                lease_expires_at
                if lease_expires_at is not None
                else (None if target in TERMINAL_STATES else current.lease_expires_at)
            ),
            updated_at=utc_now(),
        )
        self.store.jobs[current.job_id] = updated
        return updated
    def _requeue_expired_locked(self) -> None:
        now = datetime.now(timezone.utc)
        for current in list(self.store.jobs.values()):
            expiry = self._parse(current.lease_expires_at)
            if (
                current.state in {HermesJobState.CLAIMED, HermesJobState.RUNNING}
                and expiry is not None
                and expiry <= now
            ):
                self.store.jobs[current.job_id] = replace(
                    current,
                    state=HermesJobState.QUEUED,
                    revision=current.revision + 1,
                    assigned_worker_id=None,
                    lease_expires_at=None,
                    updated_at=utc_now(),
                )
@@ -0,0 +1,153 @@
 """Tenant-scoped Hermes runtime contracts."""
 from __future__ import annotations
 import json
 import uuid
 from dataclasses import dataclass, field
 from datetime import datetime, timezone
 from enum import Enum
 from typing import Any, Mapping
 def utc_now() -> str:
    return datetime.now(timezone.utc).isoformat().replace("+00:00", "Z")
 def _required(value: str, label: str) -> str:
    normalized = str(value or "").strip()
    if not normalized:
        raise ValueError(f"{label} is required")
    return normalized
 def _safe_mapping(value: Mapping[str, Any] | None) -> dict[str, Any]:
    return json.loads(json.dumps(dict(value or {}), ensure_ascii=False, sort_keys=True, default=str))
 class HermesJobState(str, Enum):
    QUEUED = "QUEUED"
    CLAIMED = "CLAIMED"
    RUNNING = "RUNNING"
    COMPLETED = "COMPLETED"
    FAILED = "FAILED"
    CANCELLED = "CANCELLED"
 ALLOWED_TRANSITIONS: Mapping[HermesJobState, frozenset[HermesJobState]] = {
    HermesJobState.QUEUED: frozenset({HermesJobState.CLAIMED, HermesJobState.CANCELLED}),
    HermesJobState.CLAIMED: frozenset({HermesJobState.RUNNING, HermesJobState.CANCELLED}),
    HermesJobState.RUNNING: frozenset(
        {HermesJobState.COMPLETED, HermesJobState.FAILED, HermesJobState.CANCELLED}
    ),
    HermesJobState.COMPLETED: frozenset(),
    HermesJobState.FAILED: frozenset(),
    HermesJobState.CANCELLED: frozenset(),
 }
 TERMINAL_STATES = frozenset(
    {HermesJobState.COMPLETED, HermesJobState.FAILED, HermesJobState.CANCELLED}
 )
@dataclass(frozen=True)
 class HermesJob:
    job_id: str
    tenant_id: str
    cell_id: str
    profile_id: str
    task_id: str
    trace_id: str
    action: str
    payload: Mapping[str, Any] = field(default_factory=dict)
    idempotency_key: str = ""
    priority: int = 100
    state: HermesJobState = HermesJobState.QUEUED
    revision: int = 1
    assigned_worker_id: str | None = None
    lease_expires_at: str | None = None
    created_at: str = field(default_factory=utc_now)
    updated_at: str = field(default_factory=utc_now)
    def __post_init__(self) -> None:
        for name in (
            "job_id", "tenant_id", "cell_id", "profile_id", "task_id",
            "trace_id", "action", "idempotency_key", "created_at", "updated_at",
        ):
            object.__setattr__(self, name, _required(getattr(self, name), name))
        object.__setattr__(self, "payload", _safe_mapping(self.payload))
        object.__setattr__(self, "state", HermesJobState(self.state))
        if not 0 <= int(self.priority) <= 1000:
            raise ValueError("priority must be between 0 and 1000")
        if self.revision < 1:
            raise ValueError("revision must be >= 1")
    @classmethod
    def create(
        cls,
        *,
        tenant_id: str,
        cell_id: str,
        profile_id: str,
        task_id: str,
        trace_id: str,
        action: str,
        payload: Mapping[str, Any] | None,
        idempotency_key: str,
        priority: int = 100,
    ) -> "HermesJob":
        return cls(
            job_id=f"hermes_{uuid.uuid4().hex}",
            tenant_id=tenant_id,
            cell_id=cell_id,
            profile_id=profile_id,
            task_id=task_id,
            trace_id=trace_id,
            action=action,
            payload=_safe_mapping(payload),
            idempotency_key=idempotency_key,
            priority=priority,
        )
@dataclass(frozen=True)
 class HermesResult:
    job_id: str
    tenant_id: str
    task_id: str
    trace_id: str
    state: HermesJobState
    worker_id: str | None = None
    output: Mapping[str, Any] | None = None
    error: Mapping[str, Any] | None = None
    started_at: str | None = None
    finished_at: str | None = None
    def __post_init__(self) -> None:
        for name in ("job_id", "tenant_id", "task_id", "trace_id"):
            object.__setattr__(self, name, _required(getattr(self, name), name))
        object.__setattr__(self, "state", HermesJobState(self.state))
        if self.output is not None:
            object.__setattr__(self, "output", _safe_mapping(self.output))
        if self.error is not None:
            object.__setattr__(self, "error", _safe_mapping(self.error))
@dataclass(frozen=True)
 class HermesHealth:
    runtime_id: str
    status: str
    version: str
    heartbeat_at: str
    active_workers: int = 0
    active_jobs: int = 0
    queued_jobs: int = 0
    tenant_partitions: int = 0
    profile_isolation_enforced: bool = True
    details: Mapping[str, Any] = field(default_factory=dict)
    def __post_init__(self) -> None:
        for name in ("runtime_id", "status", "version", "heartbeat_at"):
            object.__setattr__(self, name, _required(getattr(self, name), name))
        for name in ("active_workers", "active_jobs", "queued_jobs", "tenant_partitions"):
            if int(getattr(self, name)) < 0:
                raise ValueError(f"{name} must be >= 0")
        object.__setattr__(self, "details", _safe_mapping(self.details))
@@ -0,0 +1,155 @@
 """Hermes worker loop with lease renewal and bounded recovery."""
 from __future__ import annotations
 import threading
 from dataclasses import dataclass
 from typing import Any, Callable, Mapping
 from .adapter import HermesAdapter, HermesConflict
 from .contract import HermesJob, HermesJobState
 JobHandler = Callable[[HermesJob], Mapping[str, Any]]
@dataclass(frozen=True)
 class WorkerRunResult:
    claimed: bool
    job_id: str | None = None
    state: str | None = None
    error: str | None = None
 class HermesWorker:
    """Poll one tenant/cell/profile partition and execute jobs safely."""
    def __init__(
        self,
        adapter: HermesAdapter,
        handler: JobHandler,
        *,
        worker_id: str,
        tenant_id: str,
        cell_id: str,
        profile_id: str,
        poll_seconds: float = 1.0,
        heartbeat_seconds: float = 10.0,
    ) -> None:
        if poll_seconds <= 0 or heartbeat_seconds <= 0:
            raise ValueError("poll_seconds and heartbeat_seconds must be positive")
        for name, value in {
            "worker_id": worker_id,
            "tenant_id": tenant_id,
            "cell_id": cell_id,
            "profile_id": profile_id,
        }.items():
            if not str(value or "").strip():
                raise ValueError(f"{name} is required")
        self.adapter = adapter
        self.handler = handler
        self.worker_id = worker_id
        self.tenant_id = tenant_id
        self.cell_id = cell_id
        self.profile_id = profile_id
        self.poll_seconds = float(poll_seconds)
        self.heartbeat_seconds = float(heartbeat_seconds)
        self._stop = threading.Event()
    def stop(self) -> None:
        self._stop.set()
    def run_once(self) -> WorkerRunResult:
        job = self.adapter.claim(
            tenant_id=self.tenant_id,
            cell_id=self.cell_id,
            profile_id=self.profile_id,
            worker_id=self.worker_id,
        )
        if job is None:
            return WorkerRunResult(claimed=False)
        running = self.adapter.start(
            job.job_id,
            tenant_id=self.tenant_id,
            worker_id=self.worker_id,
            expected_revision=job.revision,
        )
        heartbeat_stop = threading.Event()
        heartbeat_thread = threading.Thread(
            target=self._heartbeat_loop,
            args=(running.job_id, heartbeat_stop),
            name=f"hermes-heartbeat-{self.worker_id}",
            daemon=True,
        )
        heartbeat_thread.start()
        output: Mapping[str, Any] | None = None
        execution_error: Exception | None = None
        try:
            output = dict(self.handler(running))
        except Exception as exc:  # runtime boundary: normalize below
            execution_error = exc
        finally:
            # Stop the lease heartbeat before reading the final revision. This
            # prevents a revision increment racing with complete/fail.
            heartbeat_stop.set()
            heartbeat_thread.join(timeout=max(self.heartbeat_seconds * 2, 1.0))
        try:
            current = self.adapter.get(running.job_id, tenant_id=self.tenant_id)
            if execution_error is None:
                result = self.adapter.complete(
                    running.job_id,
                    tenant_id=self.tenant_id,
                    worker_id=self.worker_id,
                    output=dict(output or {}),
                    expected_revision=current.revision,
                )
                return WorkerRunResult(
                    claimed=True,
                    job_id=result.job_id,
                    state=result.state.value,
                )
            result = self.adapter.fail(
                running.job_id,
                tenant_id=self.tenant_id,
                worker_id=self.worker_id,
                error={
                    "code": "worker_error",
                    "message": str(execution_error),
                },
                expected_revision=current.revision,
            )
            return WorkerRunResult(
                claimed=True,
                job_id=result.job_id,
                state=result.state.value,
                error=str(execution_error),
            )
        except HermesConflict as exc:
            return WorkerRunResult(
                claimed=True,
                job_id=running.job_id,
                state=HermesJobState.FAILED.value,
                error=f"reconciliation_conflict: {exc}",
            )
    def run_forever(self) -> None:
        while not self._stop.is_set():
            result = self.run_once()
            if not result.claimed:
                self._stop.wait(self.poll_seconds)
    def _heartbeat_loop(self, job_id: str, stop_event: threading.Event) -> None:
        while not stop_event.wait(self.heartbeat_seconds):
            try:
                self.adapter.heartbeat(
                    job_id,
                    tenant_id=self.tenant_id,
                    worker_id=self.worker_id,
                )
            except Exception:
                stop_event.set()
                return
@@ -0,0 +1,37 @@
 """AIOS-managed OpenClaw execution adapter."""
 from .adapter import (
    AdapterError,
    CommandConflict,
    CommandNotFound,
    CommandRejected,
    InMemoryCallbackSink,
    OpenClawAdapter,
 )
 from .contract import (
    CommandState,
    OpenClawCommand,
    OpenClawResult,
    RuntimeHealth,
    sign_command,
    verify_command_signature,
 )
 from .transport import HTTPRuntimeTransport, InProcessRuntimeTransport, RuntimeTransport
 __all__ = [
    "AdapterError",
    "CommandConflict",
    "CommandNotFound",
    "CommandRejected",
    "CommandState",
    "HTTPRuntimeTransport",
    "InMemoryCallbackSink",
    "InProcessRuntimeTransport",
    "OpenClawAdapter",
    "OpenClawCommand",
    "OpenClawResult",
    "RuntimeHealth",
    "RuntimeTransport",
    "sign_command",
    "verify_command_signature",
 ]
@@ -0,0 +1,228 @@
 """Signed, tenant-isolated execution adapter for OpenClaw runtimes."""
 from __future__ import annotations
 import threading
 from dataclasses import replace
 from typing import Callable, Mapping, Protocol
 from .contract import (
    CommandState,
    OpenClawCommand,
    OpenClawResult,
    RuntimeHealth,
    utc_now,
    verify_command_signature,
 )
 class AdapterError(RuntimeError):
    """Base adapter failure."""
 class CommandRejected(AdapterError):
    """A command failed signature or contract validation."""
 class CommandNotFound(AdapterError):
    """A tenant-scoped command is not visible."""
 class CommandConflict(AdapterError):
    """A command or idempotency key conflicts with an existing execution."""
 class RuntimeTransport(Protocol):
    def execute(self, command: OpenClawCommand, timeout_seconds: float) -> OpenClawResult: ...
    def cancel(self, command_id: str) -> bool: ...
    def health(self) -> RuntimeHealth: ...
 class CallbackSink(Protocol):
    def on_started(self, result: OpenClawResult) -> None: ...
    def on_completed(self, result: OpenClawResult) -> None: ...
    def on_failed(self, result: OpenClawResult) -> None: ...
 class InMemoryCallbackSink:
    def __init__(self) -> None:
        self._lock = threading.RLock()
        self.started: list[OpenClawResult] = []
        self.completed: list[OpenClawResult] = []
        self.failed: list[OpenClawResult] = []
    def on_started(self, result: OpenClawResult) -> None:
        with self._lock:
            self.started.append(result)
    def on_completed(self, result: OpenClawResult) -> None:
        with self._lock:
            self.completed.append(result)
    def on_failed(self, result: OpenClawResult) -> None:
        with self._lock:
            self.failed.append(result)
 SigningMaterialResolver = Callable[[str, str], bytes | str]
 class OpenClawAdapter:
    """Verify, de-duplicate, execute, cancel and reconcile commands."""
    def __init__(
        self,
        transport: RuntimeTransport,
        signing_material_resolver: SigningMaterialResolver,
        *,
        callback_sink: CallbackSink | None = None,
        default_timeout_seconds: float = 300.0,
    ) -> None:
        if default_timeout_seconds <= 0:
            raise ValueError("default_timeout_seconds must be positive")
        self.transport = transport
        self.signing_material_resolver = signing_material_resolver
        self.callback_sink = callback_sink or InMemoryCallbackSink()
        self.default_timeout_seconds = float(default_timeout_seconds)
        self._lock = threading.RLock()
        self._commands: dict[str, OpenClawCommand] = {}
        self._results: dict[str, OpenClawResult] = {}
        self._idempotency: dict[tuple[str, str], tuple[str, str]] = {}
    def _verify(self, command: OpenClawCommand) -> None:
        try:
            key_material = self.signing_material_resolver(command.tenant_id, command.key_id)
        except Exception as exc:
            raise CommandRejected("signing material unavailable") from exc
        if not verify_command_signature(command, key_material):
            raise CommandRejected("invalid or expired command signature")
    @staticmethod
    def _validate_result(command: OpenClawCommand, result: OpenClawResult) -> None:
        if (
            result.command_id != command.command_id
            or result.tenant_id != command.tenant_id
            or result.task_id != command.task_id
            or result.trace_id != command.trace_id
        ):
            raise AdapterError("runtime returned mismatched command identity")
        if result.state not in {
            CommandState.COMPLETED,
            CommandState.FAILED,
            CommandState.CANCELLED,
            CommandState.TIMED_OUT,
        }:
            raise AdapterError("runtime returned a non-terminal result")
    def execute(
        self,
        command: OpenClawCommand,
        *,
        timeout_seconds: float | None = None,
    ) -> OpenClawResult:
        timeout = self.default_timeout_seconds if timeout_seconds is None else float(timeout_seconds)
        if timeout <= 0:
            raise ValueError("timeout_seconds must be positive")
        self._verify(command)
        digest = command.signature
        idem_ref = (command.tenant_id, command.idempotency_key)
        with self._lock:
            prior_command = self._commands.get(command.command_id)
            if prior_command is not None:
                if prior_command.signature != command.signature:
                    raise CommandConflict("command_id reused with different content")
                return self._results[command.command_id]
            prior_idem = self._idempotency.get(idem_ref)
            if prior_idem is not None:
                prior_id, prior_digest = prior_idem
                if prior_digest != digest:
                    raise CommandConflict("idempotency key reused with different content")
                return self._results[prior_id]
            running = OpenClawResult(
                command_id=command.command_id,
                tenant_id=command.tenant_id,
                task_id=command.task_id,
                trace_id=command.trace_id,
                state=CommandState.RUNNING,
                started_at=utc_now(),
            )
            self._commands[command.command_id] = command
            self._results[command.command_id] = running
            self._idempotency[idem_ref] = (command.command_id, digest)
        self.callback_sink.on_started(running)
        try:
            result = self.transport.execute(command, timeout)
            self._validate_result(command, result)
        except TimeoutError as exc:
            result = replace(
                running,
                state=CommandState.TIMED_OUT,
                error={"code": "runtime_timeout", "message": str(exc) or "runtime timeout"},
                finished_at=utc_now(),
            )
        except Exception as exc:
            result = replace(
                running,
                state=CommandState.FAILED,
                error={"code": "runtime_error", "message": str(exc)},
                finished_at=utc_now(),
            )
        if result.started_at is None:
            result = replace(result, started_at=running.started_at)
        if result.finished_at is None:
            result = replace(result, finished_at=utc_now())
        with self._lock:
            self._results[command.command_id] = result
        if result.state == CommandState.COMPLETED:
            self.callback_sink.on_completed(result)
        else:
            self.callback_sink.on_failed(result)
        return result
    def status(self, command_id: str, *, tenant_id: str) -> OpenClawResult:
        with self._lock:
            command = self._commands.get(command_id)
            result = self._results.get(command_id)
            if command is None or result is None or command.tenant_id != tenant_id:
                raise CommandNotFound("command not found")
            return result
    def cancel(self, command_id: str, *, tenant_id: str) -> OpenClawResult:
        with self._lock:
            command = self._commands.get(command_id)
            current = self._results.get(command_id)
            if command is None or current is None or command.tenant_id != tenant_id:
                raise CommandNotFound("command not found")
            if current.state in {
                CommandState.COMPLETED,
                CommandState.FAILED,
                CommandState.CANCELLED,
                CommandState.TIMED_OUT,
                CommandState.REJECTED,
            }:
                return current
        if not self.transport.cancel(command_id):
            raise AdapterError("runtime rejected cancellation")
        cancelled = replace(
            current,
            state=CommandState.CANCELLED,
            error={"code": "cancelled", "message": "cancelled by AIOS"},
            finished_at=utc_now(),
        )
        with self._lock:
            self._results[command_id] = cancelled
        self.callback_sink.on_failed(cancelled)
        return cancelled
    def health(self) -> RuntimeHealth:
        return self.transport.health()
    def snapshot(self, *, tenant_id: str) -> Mapping[str, OpenClawResult]:
        with self._lock:
            return {
                command_id: result
                for command_id, result in self._results.items()
                if self._commands[command_id].tenant_id == tenant_id
            }
@@ -0,0 +1,97 @@
 """Audit-oriented lifecycle callbacks for OpenClaw execution."""
 from __future__ import annotations
 import json
 import threading
 from dataclasses import dataclass, field
 from typing import Any, Mapping, Protocol
 from .contract import OpenClawResult, utc_now
 class EventPublisher(Protocol):
    def publish(self, event: Mapping[str, Any]) -> None: ...
@dataclass(frozen=True)
 class ExecutionEvent:
    event_type: str
    tenant_id: str
    task_id: str
    trace_id: str
    command_id: str
    state: str
    created_at: str = field(default_factory=utc_now)
    details: Mapping[str, Any] = field(default_factory=dict)
    def as_dict(self) -> dict[str, Any]:
        return {
            "event_type": self.event_type,
            "tenant_id": self.tenant_id,
            "task_id": self.task_id,
            "trace_id": self.trace_id,
            "command_id": self.command_id,
            "state": self.state,
            "created_at": self.created_at,
            "details": json.loads(
                json.dumps(dict(self.details), ensure_ascii=False, default=str)
            ),
        }
 class InMemoryEventPublisher:
    """Thread-safe publisher used by tests and first integration slice."""
    def __init__(self) -> None:
        self._lock = threading.RLock()
        self.events: list[dict[str, Any]] = []
    def publish(self, event: Mapping[str, Any]) -> None:
        with self._lock:
            self.events.append(
                json.loads(json.dumps(dict(event), ensure_ascii=False, default=str))
            )
 class AuditCallbackSink:
    """Convert runtime callbacks into tenant-scoped audit events."""
    def __init__(self, publisher: EventPublisher) -> None:
        self.publisher = publisher
    @staticmethod
    def _event(
        event_type: str,
        result: OpenClawResult,
    ) -> ExecutionEvent:
        details: dict[str, Any] = {
            "runtime_id": result.runtime_id,
            "attempts": result.attempts,
        }
        if result.error:
            details["error"] = dict(result.error)
        if result.output:
            details["output_present"] = True
        return ExecutionEvent(
            event_type=event_type,
            tenant_id=result.tenant_id,
            task_id=result.task_id,
            trace_id=result.trace_id,
            command_id=result.command_id,
            state=result.state.value,
            details=details,
        )
    def on_started(self, result: OpenClawResult) -> None:
        self.publisher.publish(self._event("OPENCLAW_STARTED", result).as_dict())
    def on_completed(self, result: OpenClawResult) -> None:
        self.publisher.publish(self._event("OPENCLAW_COMPLETED", result).as_dict())
    def on_failed(self, result: OpenClawResult) -> None:
        event_type = {
            "CANCELLED": "OPENCLAW_CANCELLED",
            "TIMED_OUT": "OPENCLAW_TIMED_OUT",
            "REJECTED": "OPENCLAW_REJECTED",
        }.get(result.state.value, "OPENCLAW_FAILED")
        self.publisher.publish(self._event(event_type, result).as_dict())
@@ -0,0 +1,213 @@
 """Versioned AIOS to OpenClaw execution contracts."""
 from __future__ import annotations
 import hashlib
 import hmac
 import json
 import uuid
 from dataclasses import dataclass, field, replace
 from datetime import datetime, timezone
 from enum import Enum
 from typing import Any, Mapping
 CONTRACT_VERSION = "aios.openclaw.command.v1"
 DEFAULT_CLOCK_SKEW_SECONDS = 300
 def utc_now() -> str:
    return datetime.now(timezone.utc).isoformat().replace("+00:00", "Z")
 def _parse_timestamp(value: str) -> datetime:
    normalized = str(value or "").strip()
    if not normalized:
        raise ValueError("timestamp is required")
    if normalized.endswith("Z"):
        normalized = normalized[:-1] + "+00:00"
    parsed = datetime.fromisoformat(normalized)
    if parsed.tzinfo is None:
        raise ValueError("timestamp must include a timezone")
    return parsed.astimezone(timezone.utc)
 def _required(value: str, label: str) -> str:
    normalized = str(value or "").strip()
    if not normalized:
        raise ValueError(f"{label} is required")
    return normalized
 def _safe_mapping(value: Mapping[str, Any] | None) -> dict[str, Any]:
    return json.loads(json.dumps(dict(value or {}), ensure_ascii=False, sort_keys=True, default=str))
 class CommandState(str, Enum):
    ACCEPTED = "ACCEPTED"
    RUNNING = "RUNNING"
    COMPLETED = "COMPLETED"
    FAILED = "FAILED"
    CANCELLED = "CANCELLED"
    TIMED_OUT = "TIMED_OUT"
    REJECTED = "REJECTED"
@dataclass(frozen=True)
 class OpenClawCommand:
    command_id: str
    tenant_id: str
    task_id: str
    trace_id: str
    action: str
    payload: Mapping[str, Any] = field(default_factory=dict)
    idempotency_key: str = ""
    created_at: str = field(default_factory=utc_now)
    expires_at: str | None = None
    contract_version: str = CONTRACT_VERSION
    key_id: str = "default"
    signature: str = ""
    def __post_init__(self) -> None:
        for field_name in (
            "command_id", "tenant_id", "task_id", "trace_id", "action",
            "idempotency_key", "created_at", "contract_version", "key_id",
        ):
            object.__setattr__(self, field_name, _required(getattr(self, field_name), field_name))
        if self.contract_version != CONTRACT_VERSION:
            raise ValueError(f"unsupported contract_version: {self.contract_version}")
        object.__setattr__(self, "payload", _safe_mapping(self.payload))
        _parse_timestamp(self.created_at)
        if self.expires_at is not None:
            object.__setattr__(self, "expires_at", _required(self.expires_at, "expires_at"))
            _parse_timestamp(self.expires_at)
    @classmethod
    def create(
        cls,
        *,
        tenant_id: str,
        task_id: str,
        trace_id: str,
        action: str,
        payload: Mapping[str, Any] | None,
        idempotency_key: str,
        expires_at: str | None = None,
        key_id: str = "default",
        command_id: str | None = None,
    ) -> "OpenClawCommand":
        return cls(
            command_id=command_id or f"cmd_{uuid.uuid4().hex}",
            tenant_id=tenant_id,
            task_id=task_id,
            trace_id=trace_id,
            action=action,
            payload=_safe_mapping(payload),
            idempotency_key=idempotency_key,
            expires_at=expires_at,
            key_id=key_id,
        )
    def unsigned_dict(self) -> dict[str, Any]:
        return {
            "command_id": self.command_id,
            "tenant_id": self.tenant_id,
            "task_id": self.task_id,
            "trace_id": self.trace_id,
            "action": self.action,
            "payload": _safe_mapping(self.payload),
            "idempotency_key": self.idempotency_key,
            "created_at": self.created_at,
            "expires_at": self.expires_at,
            "contract_version": self.contract_version,
            "key_id": self.key_id,
        }
@dataclass(frozen=True)
 class OpenClawResult:
    command_id: str
    tenant_id: str
    task_id: str
    trace_id: str
    state: CommandState
    output: Mapping[str, Any] | None = None
    error: Mapping[str, Any] | None = None
    started_at: str | None = None
    finished_at: str | None = None
    runtime_id: str | None = None
    attempts: int = 1
    def __post_init__(self) -> None:
        for field_name in ("command_id", "tenant_id", "task_id", "trace_id"):
            object.__setattr__(self, field_name, _required(getattr(self, field_name), field_name))
        object.__setattr__(self, "state", CommandState(self.state))
        if self.output is not None:
            object.__setattr__(self, "output", _safe_mapping(self.output))
        if self.error is not None:
            object.__setattr__(self, "error", _safe_mapping(self.error))
        if self.started_at is not None:
            _parse_timestamp(self.started_at)
        if self.finished_at is not None:
            _parse_timestamp(self.finished_at)
        if self.attempts < 1:
            raise ValueError("attempts must be >= 1")
@dataclass(frozen=True)
 class RuntimeHealth:
    runtime: str
    runtime_id: str
    status: str
    version: str
    heartbeat_at: str
    active_tasks: int = 0
    workspace_isolated: bool = False
    credential_ref_configured: bool = False
    details: Mapping[str, Any] = field(default_factory=dict)
    def __post_init__(self) -> None:
        for field_name in ("runtime", "runtime_id", "status", "version", "heartbeat_at"):
            object.__setattr__(self, field_name, _required(getattr(self, field_name), field_name))
        _parse_timestamp(self.heartbeat_at)
        if self.active_tasks < 0:
            raise ValueError("active_tasks must be >= 0")
        object.__setattr__(self, "details", _safe_mapping(self.details))
 def canonical_command_bytes(command: OpenClawCommand) -> bytes:
    return json.dumps(
        command.unsigned_dict(), ensure_ascii=False, sort_keys=True, separators=(",", ":")
    ).encode("utf-8")
 def sign_command(command: OpenClawCommand, key_material: bytes | str) -> OpenClawCommand:
    key = key_material.encode("utf-8") if isinstance(key_material, str) else bytes(key_material)
    if not key:
        raise ValueError("signing material is required")
    signature = hmac.new(key, canonical_command_bytes(command), hashlib.sha256).hexdigest()
    return replace(command, signature=signature)
 def verify_command_signature(
    command: OpenClawCommand,
    key_material: bytes | str,
    *,
    now: datetime | None = None,
    clock_skew_seconds: int = DEFAULT_CLOCK_SKEW_SECONDS,
 ) -> bool:
    if not command.signature:
        return False
    key = key_material.encode("utf-8") if isinstance(key_material, str) else bytes(key_material)
    if not key:
        return False
    expected = hmac.new(key, canonical_command_bytes(command), hashlib.sha256).hexdigest()
    if not hmac.compare_digest(expected, command.signature):
        return False
    current = (now or datetime.now(timezone.utc)).astimezone(timezone.utc)
    created = _parse_timestamp(command.created_at)
    if created.timestamp() - current.timestamp() > clock_skew_seconds:
        return False
    if command.expires_at is not None:
        expires = _parse_timestamp(command.expires_at)
        if current.timestamp() - expires.timestamp() > clock_skew_seconds:
            return False
    return True
@@ -0,0 +1,103 @@
 """Health policy and self-healing decisions for OpenClaw runtimes."""
 from __future__ import annotations
 from dataclasses import dataclass, field
 from datetime import datetime, timezone
 from typing import Mapping
 from .contract import RuntimeHealth
 def _parse(value: str) -> datetime:
    normalized = value[:-1] + "+00:00" if value.endswith("Z") else value
    parsed = datetime.fromisoformat(normalized)
    if parsed.tzinfo is None:
        raise ValueError("heartbeat timestamp must include timezone")
    return parsed.astimezone(timezone.utc)
@dataclass(frozen=True)
 class HealthPolicy:
    max_heartbeat_age_seconds: int = 90
    require_workspace_isolation: bool = True
    require_credential_ref: bool = True
    accepted_statuses: frozenset[str] = field(
        default_factory=lambda: frozenset({"healthy", "ready"})
    )
    def __post_init__(self) -> None:
        if self.max_heartbeat_age_seconds < 5:
            raise ValueError("max_heartbeat_age_seconds must be >= 5")
        if not self.accepted_statuses:
            raise ValueError("accepted_statuses must not be empty")
@dataclass(frozen=True)
 class HealthAssessment:
    healthy: bool
    reasons: tuple[str, ...]
    recoverable: bool
    recommended_action: str
    details: Mapping[str, object] = field(default_factory=dict)
 class RuntimeHealthMonitor:
    """Evaluate runtime health and recommend bounded recovery actions."""
    def __init__(self, policy: HealthPolicy | None = None) -> None:
        self.policy = policy or HealthPolicy()
    def assess(
        self,
        health: RuntimeHealth,
        *,
        now: datetime | None = None,
    ) -> HealthAssessment:
        current = (now or datetime.now(timezone.utc)).astimezone(timezone.utc)
        heartbeat = _parse(health.heartbeat_at)
        age = max(0.0, current.timestamp() - heartbeat.timestamp())
        reasons: list[str] = []
        if health.status.lower() not in self.policy.accepted_statuses:
            reasons.append("runtime_status_unhealthy")
        if age > self.policy.max_heartbeat_age_seconds:
            reasons.append("heartbeat_stale")
        if self.policy.require_workspace_isolation and not health.workspace_isolated:
            reasons.append("workspace_not_isolated")
        if self.policy.require_credential_ref and not health.credential_ref_configured:
            reasons.append("credential_reference_missing")
        if not reasons:
            return HealthAssessment(
                healthy=True,
                reasons=(),
                recoverable=True,
                recommended_action="none",
                details={"heartbeat_age_seconds": age},
            )
        non_recoverable = {
            "workspace_not_isolated",
            "credential_reference_missing",
        }.intersection(reasons)
        if non_recoverable:
            action = "quarantine_runtime"
            recoverable = False
        elif "heartbeat_stale" in reasons:
            action = "restart_runtime"
            recoverable = True
        else:
            action = "drain_and_restart"
            recoverable = True
        return HealthAssessment(
            healthy=False,
            reasons=tuple(reasons),
            recoverable=recoverable,
            recommended_action=action,
            details={
                "heartbeat_age_seconds": age,
                "runtime_id": health.runtime_id,
                "active_tasks": health.active_tasks,
            },
        )
@@ -0,0 +1,214 @@
 """Runtime transport implementations for the OpenClaw adapter."""
 from __future__ import annotations
 import json
 import threading
 import urllib.error
 import urllib.parse
 import urllib.request
 from typing import Any, Callable, Mapping, Protocol
 from .contract import (
    CommandState,
    OpenClawCommand,
    OpenClawResult,
    RuntimeHealth,
    utc_now,
 )
 class RuntimeTransport(Protocol):
    def execute(self, command: OpenClawCommand, timeout_seconds: float) -> OpenClawResult: ...
    def cancel(self, command_id: str) -> bool: ...
    def health(self) -> RuntimeHealth: ...
 class TransportError(RuntimeError):
    """Runtime transport or protocol failure."""
 RuntimeHandler = Callable[[OpenClawCommand], Mapping[str, Any] | OpenClawResult]
 class InProcessRuntimeTransport:
    """Deterministic runtime used by tests and controlled local integrations."""
    def __init__(
        self,
        handler: RuntimeHandler,
        *,
        runtime_id: str = "openclaw-in-process",
        version: str = "v1",
        workspace_isolated: bool = True,
        credential_ref_configured: bool = True,
    ) -> None:
        self.handler = handler
        self.runtime_id = runtime_id
        self.version = version
        self.workspace_isolated = workspace_isolated
        self.credential_ref_configured = credential_ref_configured
        self._lock = threading.RLock()
        self._running: set[str] = set()
        self._cancelled: set[str] = set()
    def execute(self, command: OpenClawCommand, timeout_seconds: float) -> OpenClawResult:
        if timeout_seconds <= 0:
            raise ValueError("timeout_seconds must be positive")
        with self._lock:
            if command.command_id in self._cancelled:
                return OpenClawResult(
                    command_id=command.command_id,
                    tenant_id=command.tenant_id,
                    task_id=command.task_id,
                    trace_id=command.trace_id,
                    state=CommandState.CANCELLED,
                    error={"code": "cancelled", "message": "cancelled before execution"},
                    finished_at=utc_now(),
                    runtime_id=self.runtime_id,
                )
            self._running.add(command.command_id)
        try:
            raw = self.handler(command)
            if isinstance(raw, OpenClawResult):
                return raw
            if not isinstance(raw, Mapping):
                raise TransportError("runtime handler must return a mapping or OpenClawResult")
            return OpenClawResult(
                command_id=command.command_id,
                tenant_id=command.tenant_id,
                task_id=command.task_id,
                trace_id=command.trace_id,
                state=CommandState(str(raw.get("state", "COMPLETED"))),
                output=raw.get("output"),
                error=raw.get("error"),
                started_at=str(raw.get("started_at") or utc_now()),
                finished_at=str(raw.get("finished_at") or utc_now()),
                runtime_id=str(raw.get("runtime_id") or self.runtime_id),
                attempts=int(raw.get("attempts", 1)),
            )
        finally:
            with self._lock:
                self._running.discard(command.command_id)
    def cancel(self, command_id: str) -> bool:
        with self._lock:
            self._cancelled.add(command_id)
            return command_id in self._running
    def health(self) -> RuntimeHealth:
        with self._lock:
            active = len(self._running)
        return RuntimeHealth(
            runtime="openclaw",
            runtime_id=self.runtime_id,
            status="healthy",
            version=self.version,
            heartbeat_at=utc_now(),
            active_tasks=active,
            workspace_isolated=self.workspace_isolated,
            credential_ref_configured=self.credential_ref_configured,
            details={"transport": "in_process"},
        )
 class HTTPRuntimeTransport:
    """Private HTTP transport for an OpenClaw runtime service."""
    def __init__(
        self,
        base_url: str,
        *,
        runtime_id: str = "openclaw-http",
        credential_ref: str | None = None,
        max_response_bytes: int = 2_097_152,
    ) -> None:
        normalized = str(base_url).rstrip("/")
        if not normalized.startswith(("http://", "https://")):
            raise ValueError("base_url must use http or https")
        self.base_url = normalized
        self.runtime_id = runtime_id
        self.credential_ref = credential_ref
        self.max_response_bytes = int(max_response_bytes)
    def _request(
        self,
        method: str,
        path: str,
        *,
        payload: Mapping[str, Any] | None,
        timeout_seconds: float,
    ) -> Mapping[str, Any]:
        body = None
        headers = {"Accept": "application/json", "User-Agent": "AIOS-OpenClaw/1"}
        if payload is not None:
            body = json.dumps(dict(payload), separators=(",", ":")).encode("utf-8")
            headers["Content-Type"] = "application/json"
        if self.credential_ref:
            headers["X-AIOS-Credential-Ref"] = self.credential_ref
        request = urllib.request.Request(
            f"{self.base_url}{path}",
            data=body,
            method=method,
            headers=headers,
        )
        try:
            with urllib.request.urlopen(request, timeout=timeout_seconds) as response:
                raw = response.read(self.max_response_bytes + 1)
        except urllib.error.HTTPError as exc:
            raise TransportError(f"runtime returned HTTP {exc.code}") from exc
        except urllib.error.URLError as exc:
            raise TransportError(f"runtime unavailable: {exc.reason}") from exc
        if len(raw) > self.max_response_bytes:
            raise TransportError("runtime response exceeded size limit")
        decoded = json.loads(raw.decode("utf-8"))
        if not isinstance(decoded, Mapping):
            raise TransportError("runtime response must be a JSON object")
        return decoded
    @staticmethod
    def _command_payload(command: OpenClawCommand) -> dict[str, Any]:
        return {**command.unsigned_dict(), "signature": command.signature}
    def execute(self, command: OpenClawCommand, timeout_seconds: float) -> OpenClawResult:
        value = self._request(
            "POST",
            "/v1/execute",
            payload=self._command_payload(command),
            timeout_seconds=timeout_seconds,
        )
        return OpenClawResult(
            command_id=str(value["command_id"]),
            tenant_id=str(value["tenant_id"]),
            task_id=str(value["task_id"]),
            trace_id=str(value["trace_id"]),
            state=CommandState(str(value["state"])),
            output=value.get("output"),
            error=value.get("error"),
            started_at=value.get("started_at"),
            finished_at=value.get("finished_at"),
            runtime_id=str(value.get("runtime_id") or self.runtime_id),
            attempts=int(value.get("attempts", 1)),
        )
    def cancel(self, command_id: str) -> bool:
        value = self._request(
            "POST",
            f"/v1/commands/{urllib.parse.quote(command_id, safe='')}/cancel",
            payload={},
            timeout_seconds=10.0,
        )
        return bool(value.get("accepted"))
    def health(self) -> RuntimeHealth:
        value = self._request("GET", "/v1/health", payload=None, timeout_seconds=5.0)
        return RuntimeHealth(
            runtime=str(value.get("runtime") or "openclaw"),
            runtime_id=str(value.get("runtime_id") or self.runtime_id),
            status=str(value.get("status") or "unknown"),
            version=str(value.get("version") or "unknown"),
            heartbeat_at=str(value.get("heartbeat_at") or utc_now()),
            active_tasks=int(value.get("active_tasks", 0)),
            workspace_isolated=bool(value.get("workspace_isolated", False)),
            credential_ref_configured=bool(value.get("credential_ref_configured", False)),
            details=value.get("details") if isinstance(value.get("details"), Mapping) else {},
        )
@@ -0,0 +1,46 @@
 """AIOS V2 control plane and execution orchestration."""
 from .control_plane import (
    AgentRecord,
    AuditEvent,
    ControlPlaneError,
    ControlPlaneService,
    IdempotencyConflict,
    InMemoryControlPlaneStore,
    InvalidTransition,
    NotFound,
    PolicyDenied,
    RequestContext,
    RevisionConflict,
    TaskRecord,
    TaskState,
    TenantIsolationError,
    TenantRecord,
    Unauthorized,
 )
 from .execution_tracking import ExecutionEvidence, ExecutionTracker
 from .orchestrator import DispatchRecord, ExecutionOrchestrator, OrchestrationError
 __all__ = [
    "AgentRecord",
    "AuditEvent",
    "ControlPlaneError",
    "ControlPlaneService",
    "DispatchRecord",
    "ExecutionEvidence",
    "ExecutionOrchestrator",
    "ExecutionTracker",
    "IdempotencyConflict",
    "InMemoryControlPlaneStore",
    "InvalidTransition",
    "NotFound",
    "OrchestrationError",
    "PolicyDenied",
    "RequestContext",
    "RevisionConflict",
    "TaskRecord",
    "TaskState",
    "TenantIsolationError",
    "TenantRecord",
    "Unauthorized",
 ]
@@ -0,0 +1,357 @@
 """Production-candidate AIOS V2 application entrypoint.
 This process owns the public control-plane API and read-only web console. It
 loads durable control-plane state, dispatches approved work through runtime
 adapters, exposes tenant-scoped execution evidence and sanitized GitOps status,
 and never exposes runtime credentials to the browser.
 """
 from __future__ import annotations
 import dataclasses
 import json
 import os
 import stat
 import urllib.parse
 from enum import Enum
 from http import HTTPStatus
 from http.server import BaseHTTPRequestHandler, ThreadingHTTPServer
 from pathlib import Path
 from typing import Any, Mapping
 from aios_hermes import HermesAdapter, HermesJobState, HermesResult
 from aios_openclaw import HTTPRuntimeTransport, OpenClawAdapter
 from aios_openclaw.callback import AuditCallbackSink, EventPublisher
 from .console import CONSOLE_HTML
 from .control_plane import AuditEvent, RequestContext
 from .execution_tracking import ExecutionTracker
 from .gateway import GatewayApplication, error_response
 from .gitops_status import collect_gitops_status
 from .orchestrator import ExecutionOrchestrator
 from .persistence import DurableControlPlaneService, SQLiteSnapshotRepository
 def _json_value(value: Any) -> Any:
    if dataclasses.is_dataclass(value):
        return {
            item.name: _json_value(getattr(value, item.name))
            for item in dataclasses.fields(value)
        }
    if isinstance(value, Enum):
        return value.value
    if isinstance(value, Mapping):
        return {str(key): _json_value(item) for key, item in value.items()}
    if isinstance(value, (list, tuple, set, frozenset)):
        return [_json_value(item) for item in value]
    return value
 class FileSigningMaterialResolver:
    """Resolve runtime signing material from a root-owned file reference."""
    def __init__(self, path: str | Path) -> None:
        self.path = Path(path).resolve()
    def __call__(self, tenant_id: str, key_id: str) -> bytes:
        del tenant_id, key_id
        metadata = self.path.stat()
        if not stat.S_ISREG(metadata.st_mode):
            raise RuntimeError("signing material reference is not a regular file")
        if metadata.st_mode & 0o077:
            raise RuntimeError("signing material file permissions must be 0600 or stricter")
        value = self.path.read_bytes().strip()
        if len(value) < 32:
            raise RuntimeError("signing material is too short")
        return value
 class ControlPlaneEventPublisher(EventPublisher):
    """Append OpenClaw runtime events to the tenant audit stream."""
    def __init__(self, service: DurableControlPlaneService) -> None:
        self.service = service
    def publish(self, event: Mapping[str, Any]) -> None:
        tenant_id = str(event.get("tenant_id") or "").strip()
        trace_id = str(event.get("trace_id") or "").strip()
        command_id = str(event.get("command_id") or "").strip()
        event_type = str(event.get("event_type") or "OPENCLAW_EVENT").strip()
        if not tenant_id or not trace_id or not command_id:
            raise ValueError("runtime audit event is missing tenant identity")
        state_value = str(event.get("state") or "unknown").lower()
        outcome = "allowed" if state_value == "completed" else state_value
        audit = AuditEvent(
            event_id=f"runtime_{command_id}_{len(self.service.store.audit_events) + 1}",
            tenant_id=tenant_id,
            principal_id="openclaw-runtime",
            trace_id=trace_id,
            action=event_type,
            resource_type="openclaw_command",
            resource_id=command_id,
            outcome=outcome,
            policy_version="runtime-v1",
            details=dict(event.get("details") or {}),
        )
        with self.service.store.lock:
            self.service.store.audit_events.append(audit)
            self.service.repository.save(self.service._snapshot())
 class CandidateApplication:
    """Compose durable control plane, runtimes, evidence and console routes."""
    def __init__(
        self,
        gateway: GatewayApplication,
        orchestrator: ExecutionOrchestrator,
        tracker: ExecutionTracker,
        hermes: HermesAdapter,
        openclaw: OpenClawAdapter | None,
    ) -> None:
        self.gateway = gateway
        self.orchestrator = orchestrator
        self.tracker = tracker
        self.hermes = hermes
        self.openclaw = openclaw
    def context(self, headers: Mapping[str, str]) -> RequestContext:
        return self.gateway.context_from_headers(headers)
    def handle_json(
        self,
        method: str,
        raw_path: str,
        headers: Mapping[str, str],
        body: Any,
    ) -> tuple[int, Mapping[str, Any]]:
        parsed = urllib.parse.urlparse(raw_path)
        path = parsed.path.rstrip("/") or "/"
        method = method.upper()
        if method == "GET" and path == "/api/v2/gitops":
            context = self.context(headers)
            return HTTPStatus.OK, {
                "tenant_id": context.tenant_id,
                "gitops": collect_gitops_status().as_dict(),
            }
        if method == "GET" and path == "/api/v2/runtime":
            context = self.context(headers)
            runtimes: list[dict[str, Any]] = []
            if self.openclaw is None:
                runtimes.append(
                    {
                        "name": "openclaw",
                        "status": "disabled",
                        "details": {"reason": "runtime_not_configured"},
                    }
                )
            else:
                try:
                    health = self.openclaw.health()
                    runtimes.append(
                        {
                            "name": "openclaw",
                            "status": health.status,
                            "details": _json_value(health),
                        }
                    )
                except Exception as exc:
                    runtimes.append(
                        {
                            "name": "openclaw",
                            "status": "unavailable",
                            "details": {"message": str(exc)},
                        }
                    )
            hermes_health = self.hermes.health()
            runtimes.append(
                {
                    "name": "hermes",
                    "status": hermes_health.status,
                    "details": _json_value(hermes_health),
                }
            )
            return HTTPStatus.OK, {
                "tenant_id": context.tenant_id,
                "runtimes": runtimes,
            }
        if method == "GET" and path == "/api/v2/executions":
            context = self.context(headers)
            return HTTPStatus.OK, {
                "executions": _json_value(
                    self.tracker.list(tenant_id=context.tenant_id)
                )
            }
        dispatch_match = __import__("re").fullmatch(
            r"/api/v2/tasks/([^/]+)/dispatch",
            path,
        )
        if method == "POST" and dispatch_match:
            context = self.context(headers)
            task_id = urllib.parse.unquote(dispatch_match.group(1))
            payload = body if isinstance(body, Mapping) else {}
            timeout = payload.get("timeout_seconds")
            record = self.orchestrator.dispatch(
                context,
                task_id=task_id,
                timeout_seconds=float(timeout) if timeout is not None else None,
            )
            return HTTPStatus.ACCEPTED, {"dispatch": _json_value(record)}
        if method == "POST" and path == "/api/v2/hermes/results/reconcile":
            context = self.context(headers)
            if not isinstance(body, Mapping):
                raise ValueError("JSON body must be an object")
            result = HermesResult(
                job_id=str(body["job_id"]),
                tenant_id=str(body["tenant_id"]),
                task_id=str(body["task_id"]),
                trace_id=str(body["trace_id"]),
                state=HermesJobState(str(body["state"])),
                worker_id=(
                    str(body["worker_id"])
                    if body.get("worker_id") is not None
                    else None
                ),
                output=body.get("output"),
                error=body.get("error"),
                started_at=body.get("started_at"),
                finished_at=body.get("finished_at"),
            )
            task = self.orchestrator.reconcile_hermes(context, result)
            return HTTPStatus.OK, {"task": _json_value(task)}
        return self.gateway.handle(method, raw_path, headers, body)
 class CandidateRequestHandler(BaseHTTPRequestHandler):
    server_version = "AIOSV2Candidate/1.0"
    max_body_bytes = 1_048_576
    @property
    def application(self) -> CandidateApplication:
        return self.server.application  # type: ignore[attr-defined]
    def _read_body(self) -> Any:
        length = int(self.headers.get("Content-Length", "0") or "0")
        if length < 0 or length > self.max_body_bytes:
            raise ValueError("request body is too large")
        if length == 0:
            return {}
        content_type = self.headers.get("Content-Type", "").split(";", 1)[0].strip()
        if content_type != "application/json":
            raise ValueError("Content-Type must be application/json")
        return json.loads(self.rfile.read(length).decode("utf-8"))
    def _security_headers(self) -> None:
        self.send_header("Cache-Control", "no-store")
        self.send_header("X-Content-Type-Options", "nosniff")
        self.send_header("X-Frame-Options", "DENY")
        self.send_header("Referrer-Policy", "no-referrer")
        self.send_header(
            "Content-Security-Policy",
            "default-src 'self'; style-src 'unsafe-inline'; script-src 'unsafe-inline'; connect-src 'self'",
        )
    def _send_json(self, status_code: int, payload: Mapping[str, Any]) -> None:
        encoded = json.dumps(
            _json_value(payload),
            ensure_ascii=False,
            separators=(",", ":"),
        ).encode("utf-8")
        self.send_response(int(status_code))
        self.send_header("Content-Type", "application/json; charset=utf-8")
        self.send_header("Content-Length", str(len(encoded)))
        self._security_headers()
        self.end_headers()
        self.wfile.write(encoded)
    def _send_console(self) -> None:
        encoded = CONSOLE_HTML.encode("utf-8")
        self.send_response(HTTPStatus.OK)
        self.send_header("Content-Type", "text/html; charset=utf-8")
        self.send_header("Content-Length", str(len(encoded)))
        self._security_headers()
        self.end_headers()
        self.wfile.write(encoded)
    def _dispatch(self, method: str) -> None:
        path = urllib.parse.urlparse(self.path).path.rstrip("/") or "/"
        if method == "GET" and path in {"/", "/console"}:
            self._send_console()
            return
        try:
            body = self._read_body() if method == "POST" else {}
            status_code, payload = self.application.handle_json(
                method,
                self.path,
                {key: value for key, value in self.headers.items()},
                body,
            )
        except BaseException as exc:
            status_code, payload = error_response(exc)
        self._send_json(status_code, payload)
    def do_GET(self) -> None:  # noqa: N802
        self._dispatch("GET")
    def do_POST(self) -> None:  # noqa: N802
        self._dispatch("POST")
 class CandidateHTTPServer(ThreadingHTTPServer):
    daemon_threads = True
    allow_reuse_address = True
    def __init__(self, address: tuple[str, int], application: CandidateApplication) -> None:
        self.application = application
        super().__init__(address, CandidateRequestHandler)
 def build_application() -> CandidateApplication:
    state_dir = Path(os.getenv("AIOS_V2_STATE_DIR", "/var/lib/aios-v2")).resolve()
    state_dir.mkdir(parents=True, exist_ok=True)
    repository = SQLiteSnapshotRepository(state_dir / "control-plane.sqlite3")
    service = DurableControlPlaneService(repository)
    tracker = ExecutionTracker(state_dir / "executions")
    hermes = HermesAdapter()
    openclaw_url = os.getenv("AIOS_OPENCLAW_URL", "").strip()
    material_file = os.getenv("AIOS_OPENCLAW_SIGNING_MATERIAL_FILE", "").strip()
    openclaw = None
    resolver = None
    if openclaw_url and material_file:
        resolver = FileSigningMaterialResolver(material_file)
        transport = HTTPRuntimeTransport(
            openclaw_url,
            credential_ref=os.getenv("AIOS_OPENCLAW_CREDENTIAL_REF") or None,
        )
        openclaw = OpenClawAdapter(
            transport,
            resolver,
            callback_sink=AuditCallbackSink(ControlPlaneEventPublisher(service)),
        )
    gateway = GatewayApplication(service)
    orchestrator = ExecutionOrchestrator(
        service,
        openclaw=openclaw,
        hermes=hermes,
        openclaw_signing_material_resolver=resolver,
    )
    return CandidateApplication(gateway, orchestrator, tracker, hermes, openclaw)
 def main() -> None:
    host = os.getenv("AIOS_V2_HOST", "127.0.0.1")
    port = int(os.getenv("AIOS_V2_PORT", "18920"))
    server = CandidateHTTPServer((host, port), build_application())
    print(f"AIOS V2 candidate listening on {host}:{port}")
    server.serve_forever()
 if __name__ == "__main__":
    main()
@@ -0,0 +1,69 @@
 """Dependency-free AIOS V2 web console assets.
 Authentication and tenant identity are injected by the trusted reverse proxy.
 The browser never selects a tenant and never calls OpenClaw or Hermes directly.
 """
 CONSOLE_HTML = r'''<!doctype html>
 <html lang="en">
 <head>
  <meta charset="utf-8">
  <meta name="viewport" content="width=device-width,initial-scale=1">
  <meta name="referrer" content="no-referrer">
  <title>AIOS V2 Console</title>
  <style>
    :root{font-family:Inter,ui-sans-serif,system-ui,sans-serif;color:#152033;background:#f4f7fb}
    *{box-sizing:border-box} body{margin:0} header{background:#111827;color:white;padding:18px 24px;display:flex;justify-content:space-between;align-items:center}
    header h1{font-size:20px;margin:0} header span{font-size:12px;color:#cbd5e1}
    main{padding:24px;max-width:1440px;margin:auto}.notice{background:#fff7ed;border:1px solid #fed7aa;padding:12px 14px;border-radius:10px;margin-bottom:18px;font-size:13px}
    .cards{display:grid;grid-template-columns:repeat(auto-fit,minmax(190px,1fr));gap:14px;margin-bottom:20px}.card,.panel{background:white;border:1px solid #e5e7eb;border-radius:12px;box-shadow:0 1px 2px rgba(0,0,0,.04)}
    .card{padding:16px}.card b{display:block;font-size:28px;margin-top:8px}.card small{color:#64748b}.grid{display:grid;grid-template-columns:repeat(auto-fit,minmax(420px,1fr));gap:16px}.panel{padding:16px;overflow:auto}.panel h2{font-size:16px;margin:0 0 12px}
    table{width:100%;border-collapse:collapse;font-size:12px}th,td{text-align:left;border-bottom:1px solid #edf0f4;padding:8px;vertical-align:top}th{color:#64748b;font-weight:600}.state{font-weight:700}.ok{color:#047857}.bad{color:#b91c1c}.muted{color:#64748b}
    button{border:0;border-radius:8px;padding:9px 13px;background:#2563eb;color:white;cursor:pointer}button:disabled{opacity:.5}.toolbar{display:flex;gap:8px;align-items:center;margin-bottom:16px}.error{color:#b91c1c;white-space:pre-wrap}
  </style>
 </head>
 <body>
 <header><h1>AIOS V2 Control Plane</h1><span id="updated">Not loaded</span></header>
 <main>
  <div class="notice">Read-only candidate console. Identity and tenant context must be injected by the trusted reverse proxy. The console never calls OpenClaw or Hermes directly.</div>
  <div class="toolbar"><button id="refresh">Refresh</button><span id="error" class="error"></span></div>
  <section class="cards">
    <div class="card"><small>Agents</small><b id="agentCount">—</b></div>
    <div class="card"><small>Tasks</small><b id="taskCount">—</b></div>
    <div class="card"><small>Audit events</small><b id="auditCount">—</b></div>
    <div class="card"><small>Executions</small><b id="executionCount">—</b></div>
  </section>
  <section class="grid">
    <div class="panel"><h2>Runtime health</h2><div id="runtime"></div></div>
    <div class="panel"><h2>Tasks</h2><div id="tasks"></div></div>
    <div class="panel"><h2>Agents</h2><div id="agents"></div></div>
    <div class="panel"><h2>Execution evidence</h2><div id="executions"></div></div>
    <div class="panel"><h2>Audit</h2><div id="audit"></div></div>
  </section>
 </main>
 <script>
 const esc=v=>String(v??'').replace(/[&<>"']/g,c=>({'&':'&amp;','<':'&lt;','>':'&gt;','"':'&quot;',"'":'&#39;'}[c]));
 const table=(cols,rows)=>`<table><thead><tr>${cols.map(c=>`<th>${esc(c[0])}</th>`).join('')}</tr></thead><tbody>${rows.map(r=>`<tr>${cols.map(c=>`<td>${esc(c[1](r))}</td>`).join('')}</tr>`).join('')}</tbody></table>`;
 async function api(path){const r=await fetch(path,{credentials:'same-origin',headers:{'Accept':'application/json'}});const data=await r.json().catch(()=>({}));if(!r.ok)throw new Error(`${r.status} ${data.message||data.error||path}`);return data}
 async function refresh(){
  const btn=document.getElementById('refresh');btn.disabled=true;document.getElementById('error').textContent='';
  try{
    const [agents,tasks,audit,runtime,executions]=await Promise.all([
      api('/api/v2/agents'),api('/api/v2/tasks'),api('/api/v2/audit'),api('/api/v2/runtime'),api('/api/v2/executions')
    ]);
    document.getElementById('agentCount').textContent=agents.agents.length;
    document.getElementById('taskCount').textContent=tasks.tasks.length;
    document.getElementById('auditCount').textContent=audit.events.length;
    document.getElementById('executionCount').textContent=executions.executions.length;
    document.getElementById('agents').innerHTML=table([['ID',x=>x.agent_id],['Runtime',x=>x.runtime],['Enabled',x=>x.enabled]],agents.agents);
    document.getElementById('tasks').innerHTML=table([['Task',x=>x.task_id],['Action',x=>x.action],['State',x=>x.state],['Rev',x=>x.revision]],tasks.tasks);
    document.getElementById('audit').innerHTML=table([['Time',x=>x.created_at],['Action',x=>x.action],['Outcome',x=>x.outcome],['Resource',x=>x.resource_id]],audit.events.slice(-100).reverse());
    document.getElementById('executions').innerHTML=table([['Executor',x=>x.executor],['Task',x=>x.task_id],['Status',x=>x.status],['Progress',x=>`${x.progress_completed}/${x.progress_total}`],['Commit',x=>x.last_commit||'']],executions.executions);
    document.getElementById('runtime').innerHTML=table([['Runtime',x=>x.name],['Status',x=>x.status],['Details',x=>JSON.stringify(x.details)]],runtime.runtimes);
    document.getElementById('updated').textContent=`Updated ${new Date().toLocaleString()}`;
  }catch(e){document.getElementById('error').textContent=e.message||String(e)}finally{btn.disabled=false}
 }
 document.getElementById('refresh').addEventListener('click',refresh);refresh();setInterval(refresh,30000);
 </script>
 </body>
 </html>'''
@@ -0,0 +1,627 @@
 """Tenant-isolated AIOS V2 control-plane domain service.
 This module deliberately contains no OpenClaw or Hermes implementation details.
 It owns authoritative tenant, agent, task-lifecycle, idempotency and audit state.
 Runtime adapters consume approved tasks through a separate boundary.
 """
 from __future__ import annotations
 import copy
 import hashlib
 import json
 import threading
 import uuid
 from dataclasses import dataclass, field
 from datetime import datetime, timezone
 from enum import Enum
 from typing import Any, Callable, Iterable, Mapping
 def _utc_now() -> str:
    return datetime.now(timezone.utc).isoformat().replace("+00:00", "Z")
 def _new_id(prefix: str) -> str:
    return f"{prefix}_{uuid.uuid4().hex}"
 def _canonical_digest(value: Mapping[str, Any]) -> str:
    encoded = json.dumps(
        value,
        ensure_ascii=False,
        sort_keys=True,
        separators=(",", ":"),
        default=str,
    ).encode("utf-8")
    return hashlib.sha256(encoded).hexdigest()
 class ControlPlaneError(RuntimeError):
    """Base error returned by the control-plane domain."""
 class Unauthorized(ControlPlaneError):
    """The caller is not authenticated or lacks the required role."""
 class TenantIsolationError(ControlPlaneError):
    """A request attempted to cross a tenant boundary."""
 class NotFound(ControlPlaneError):
    """A tenant-scoped resource does not exist or is not visible."""
 class InvalidTransition(ControlPlaneError):
    """A task state transition violates the lifecycle contract."""
 class IdempotencyConflict(ControlPlaneError):
    """An idempotency key was reused for a different command."""
 class RevisionConflict(ControlPlaneError):
    """Optimistic task revision does not match current state."""
 class PolicyDenied(ControlPlaneError):
    """The policy gate denied a control-plane operation."""
@dataclass(frozen=True)
 class RequestContext:
    """Authenticated request identity propagated across all AIOS layers."""
    tenant_id: str
    principal_id: str
    trace_id: str
    roles: frozenset[str] = field(default_factory=frozenset)
    session_id: str | None = None
    policy_version: str = "v1"
    def __post_init__(self) -> None:
        for name in ("tenant_id", "principal_id", "trace_id"):
            if not str(getattr(self, name, "")).strip():
                raise ValueError(f"{name} is required")
    @classmethod
    def create(
        cls,
        *,
        tenant_id: str,
        principal_id: str,
        roles: Iterable[str] = (),
        trace_id: str | None = None,
        session_id: str | None = None,
        policy_version: str = "v1",
    ) -> "RequestContext":
        return cls(
            tenant_id=tenant_id.strip(),
            principal_id=principal_id.strip(),
            trace_id=trace_id or _new_id("trace"),
            roles=frozenset(str(role).strip() for role in roles if str(role).strip()),
            session_id=session_id,
            policy_version=policy_version,
        )
@dataclass(frozen=True)
 class TenantRecord:
    tenant_id: str
    display_name: str
    status: str = "active"
    metadata: Mapping[str, Any] = field(default_factory=dict)
    created_at: str = field(default_factory=_utc_now)
    updated_at: str = field(default_factory=_utc_now)
@dataclass(frozen=True)
 class AgentRecord:
    agent_id: str
    tenant_id: str
    display_name: str
    runtime: str
    capabilities: tuple[str, ...] = field(default_factory=tuple)
    enabled: bool = True
    metadata: Mapping[str, Any] = field(default_factory=dict)
    created_at: str = field(default_factory=_utc_now)
    updated_at: str = field(default_factory=_utc_now)
 class TaskState(str, Enum):
    CREATED = "CREATED"
    APPROVED = "APPROVED"
    RUNNING = "RUNNING"
    COMPLETED = "COMPLETED"
    FAILED = "FAILED"
    CANCELLED = "CANCELLED"
 TERMINAL_TASK_STATES = frozenset(
    {TaskState.COMPLETED, TaskState.FAILED, TaskState.CANCELLED}
 )
 ALLOWED_TASK_TRANSITIONS: Mapping[TaskState, frozenset[TaskState]] = {
    TaskState.CREATED: frozenset({TaskState.APPROVED, TaskState.CANCELLED}),
    TaskState.APPROVED: frozenset({TaskState.RUNNING, TaskState.CANCELLED}),
    TaskState.RUNNING: frozenset(
        {TaskState.COMPLETED, TaskState.FAILED, TaskState.CANCELLED}
    ),
    TaskState.COMPLETED: frozenset(),
    TaskState.FAILED: frozenset(),
    TaskState.CANCELLED: frozenset(),
 }
@dataclass(frozen=True)
 class TaskRecord:
    task_id: str
    tenant_id: str
    agent_id: str
    action: str
    payload: Mapping[str, Any]
    state: TaskState = TaskState.CREATED
    revision: int = 1
    idempotency_key: str | None = None
    request_digest: str = ""
    result: Mapping[str, Any] | None = None
    error: Mapping[str, Any] | None = None
    created_by: str = ""
    approved_by: str | None = None
    created_at: str = field(default_factory=_utc_now)
    updated_at: str = field(default_factory=_utc_now)
@dataclass(frozen=True)
 class AuditEvent:
    event_id: str
    tenant_id: str
    principal_id: str
    trace_id: str
    action: str
    resource_type: str
    resource_id: str
    outcome: str
    policy_version: str
    details: Mapping[str, Any] = field(default_factory=dict)
    created_at: str = field(default_factory=_utc_now)
@dataclass(frozen=True)
 class PolicyDecision:
    allowed: bool
    reason: str
    policy_version: str
 PolicyEvaluator = Callable[[RequestContext, str, Mapping[str, Any]], PolicyDecision]
 class InMemoryControlPlaneStore:
    """Thread-safe reference store used by tests and the first gateway slice.
    Production backends must preserve the same tenant-scoped interface and add
    database row-level security, migrations, backup and HA evidence.
    """
    def __init__(self) -> None:
        self.lock = threading.RLock()
        self.tenants: dict[str, TenantRecord] = {}
        self.agents: dict[str, AgentRecord] = {}
        self.tasks: dict[str, TaskRecord] = {}
        self.idempotency: dict[tuple[str, str], tuple[str, str]] = {}
        self.audit_events: list[AuditEvent] = []
 class ControlPlaneService:
    """Authoritative tenant, agent and task lifecycle service."""
    def __init__(
        self,
        store: InMemoryControlPlaneStore | None = None,
        *,
        policy_evaluator: PolicyEvaluator | None = None,
    ) -> None:
        self.store = store or InMemoryControlPlaneStore()
        self.policy_evaluator = policy_evaluator or self._default_policy
    @staticmethod
    def _default_policy(
        context: RequestContext,
        action: str,
        resource: Mapping[str, Any],
    ) -> PolicyDecision:
        del resource
        role_requirements = {
            "tenant.register": {"platform_admin"},
            "tenant.read": {"tenant_admin", "operator", "auditor", "platform_admin"},
            "agent.register": {"tenant_admin", "operator", "platform_admin"},
            "agent.read": {"tenant_admin", "operator", "auditor", "platform_admin"},
            "task.create": {"tenant_admin", "operator", "platform_admin"},
            "task.read": {"tenant_admin", "operator", "auditor", "platform_admin"},
            "task.approve": {"tenant_admin", "approver", "platform_admin"},
            "task.start": {"runtime", "operator", "platform_admin"},
            "task.complete": {"runtime", "operator", "platform_admin"},
            "task.fail": {"runtime", "operator", "platform_admin"},
            "task.cancel": {"tenant_admin", "operator", "approver", "platform_admin"},
            "audit.read": {"tenant_admin", "auditor", "platform_admin"},
        }
        required = role_requirements.get(action, {"platform_admin"})
        allowed = bool(context.roles.intersection(required))
        return PolicyDecision(
            allowed=allowed,
            reason="role_allowed" if allowed else "required_role_missing",
            policy_version=context.policy_version,
        )
    def _require_context(self, context: RequestContext) -> None:
        if not context.principal_id or not context.tenant_id or not context.trace_id:
            raise Unauthorized("authenticated tenant context is required")
    def _authorize(
        self,
        context: RequestContext,
        action: str,
        resource: Mapping[str, Any],
    ) -> PolicyDecision:
        self._require_context(context)
        decision = self.policy_evaluator(context, action, resource)
        if not decision.allowed:
            self._audit(
                context,
                action=action,
                resource_type=str(resource.get("resource_type", "policy")),
                resource_id=str(resource.get("resource_id", "unknown")),
                outcome="denied",
                details={"reason": decision.reason},
            )
            raise PolicyDenied(f"{action} denied: {decision.reason}")
        return decision
    @staticmethod
    def _assert_same_tenant(context: RequestContext, tenant_id: str) -> None:
        if context.tenant_id != tenant_id and "platform_admin" not in context.roles:
            raise TenantIsolationError("cross-tenant access denied")
    def _audit(
        self,
        context: RequestContext,
        *,
        action: str,
        resource_type: str,
        resource_id: str,
        outcome: str,
        details: Mapping[str, Any] | None = None,
    ) -> AuditEvent:
        event = AuditEvent(
            event_id=_new_id("audit"),
            tenant_id=context.tenant_id,
            principal_id=context.principal_id,
            trace_id=context.trace_id,
            action=action,
            resource_type=resource_type,
            resource_id=resource_id,
            outcome=outcome,
            policy_version=context.policy_version,
            details=copy.deepcopy(dict(details or {})),
        )
        self.store.audit_events.append(event)
        return event
    def register_tenant(
        self,
        context: RequestContext,
        *,
        tenant_id: str,
        display_name: str,
        metadata: Mapping[str, Any] | None = None,
    ) -> TenantRecord:
        tenant_id = tenant_id.strip()
        if not tenant_id or not display_name.strip():
            raise ValueError("tenant_id and display_name are required")
        self._authorize(
            context,
            "tenant.register",
            {"resource_type": "tenant", "resource_id": tenant_id},
        )
        self._assert_same_tenant(context, tenant_id)
        with self.store.lock:
            if tenant_id in self.store.tenants:
                raise ControlPlaneError(f"tenant already exists: {tenant_id}")
            tenant = TenantRecord(
                tenant_id=tenant_id,
                display_name=display_name.strip(),
                metadata=copy.deepcopy(dict(metadata or {})),
            )
            self.store.tenants[tenant_id] = tenant
            self._audit(
                context,
                action="tenant.register",
                resource_type="tenant",
                resource_id=tenant_id,
                outcome="allowed",
            )
            return tenant
    def get_tenant(self, context: RequestContext, tenant_id: str | None = None) -> TenantRecord:
        tenant_id = tenant_id or context.tenant_id
        self._authorize(
            context,
            "tenant.read",
            {"resource_type": "tenant", "resource_id": tenant_id},
        )
        self._assert_same_tenant(context, tenant_id)
        with self.store.lock:
            tenant = self.store.tenants.get(tenant_id)
            if tenant is None:
                raise NotFound("tenant not found")
            return tenant
    def register_agent(
        self,
        context: RequestContext,
        *,
        agent_id: str,
        display_name: str,
        runtime: str,
        capabilities: Iterable[str] = (),
        metadata: Mapping[str, Any] | None = None,
    ) -> AgentRecord:
        self._authorize(
            context,
            "agent.register",
            {"resource_type": "agent", "resource_id": agent_id},
        )
        tenant = self.get_tenant(context)
        if tenant.status != "active":
            raise ControlPlaneError("tenant is not active")
        normalized_capabilities = tuple(
            sorted({str(item).strip() for item in capabilities if str(item).strip()})
        )
        if not agent_id.strip() or not display_name.strip() or not runtime.strip():
            raise ValueError("agent_id, display_name and runtime are required")
        with self.store.lock:
            existing = self.store.agents.get(agent_id)
            if existing is not None and existing.tenant_id != context.tenant_id:
                raise TenantIsolationError("agent id belongs to another tenant")
            if existing is not None:
                raise ControlPlaneError(f"agent already exists: {agent_id}")
            agent = AgentRecord(
                agent_id=agent_id.strip(),
                tenant_id=context.tenant_id,
                display_name=display_name.strip(),
                runtime=runtime.strip(),
                capabilities=normalized_capabilities,
                metadata=copy.deepcopy(dict(metadata or {})),
            )
            self.store.agents[agent.agent_id] = agent
            self._audit(
                context,
                action="agent.register",
                resource_type="agent",
                resource_id=agent.agent_id,
                outcome="allowed",
                details={"runtime": agent.runtime},
            )
            return agent
    def get_agent(self, context: RequestContext, agent_id: str) -> AgentRecord:
        self._authorize(
            context,
            "agent.read",
            {"resource_type": "agent", "resource_id": agent_id},
        )
        with self.store.lock:
            agent = self.store.agents.get(agent_id)
            if agent is None:
                raise NotFound("agent not found")
            self._assert_same_tenant(context, agent.tenant_id)
            return agent
    def list_agents(self, context: RequestContext) -> list[AgentRecord]:
        self._authorize(
            context,
            "agent.read",
            {"resource_type": "agent_collection", "resource_id": context.tenant_id},
        )
        with self.store.lock:
            return [
                agent
                for agent in self.store.agents.values()
                if agent.tenant_id == context.tenant_id
            ]
    def create_task(
        self,
        context: RequestContext,
        *,
        agent_id: str,
        action: str,
        payload: Mapping[str, Any] | None = None,
        idempotency_key: str | None = None,
    ) -> TaskRecord:
        self._authorize(
            context,
            "task.create",
            {"resource_type": "task", "resource_id": idempotency_key or "new"},
        )
        agent = self.get_agent(context, agent_id)
        if not agent.enabled:
            raise ControlPlaneError("agent is disabled")
        if action not in agent.capabilities and "*" not in agent.capabilities:
            raise PolicyDenied(f"agent does not advertise capability: {action}")
        normalized_payload = copy.deepcopy(dict(payload or {}))
        request_digest = _canonical_digest(
            {"agent_id": agent_id, "action": action, "payload": normalized_payload}
        )
        normalized_key = idempotency_key.strip() if idempotency_key else None
        with self.store.lock:
            if normalized_key:
                idempotency_ref = (context.tenant_id, normalized_key)
                existing = self.store.idempotency.get(idempotency_ref)
                if existing is not None:
                    existing_task_id, existing_digest = existing
                    if existing_digest != request_digest:
                        raise IdempotencyConflict(
                            "idempotency key was reused with a different request"
                        )
                    return self.store.tasks[existing_task_id]
            task = TaskRecord(
                task_id=_new_id("task"),
                tenant_id=context.tenant_id,
                agent_id=agent_id,
                action=action,
                payload=normalized_payload,
                idempotency_key=normalized_key,
                request_digest=request_digest,
                created_by=context.principal_id,
            )
            self.store.tasks[task.task_id] = task
            if normalized_key:
                self.store.idempotency[(context.tenant_id, normalized_key)] = (
                    task.task_id,
                    request_digest,
                )
            self._audit(
                context,
                action="task.create",
                resource_type="task",
                resource_id=task.task_id,
                outcome="allowed",
                details={"agent_id": agent_id, "action": action},
            )
            return task
    def get_task(self, context: RequestContext, task_id: str) -> TaskRecord:
        self._authorize(
            context,
            "task.read",
            {"resource_type": "task", "resource_id": task_id},
        )
        with self.store.lock:
            task = self.store.tasks.get(task_id)
            if task is None:
                raise NotFound("task not found")
            self._assert_same_tenant(context, task.tenant_id)
            return task
    def list_tasks(
        self,
        context: RequestContext,
        *,
        states: Iterable[TaskState | str] | None = None,
    ) -> list[TaskRecord]:
        self._authorize(
            context,
            "task.read",
            {"resource_type": "task_collection", "resource_id": context.tenant_id},
        )
        state_filter = (
            {TaskState(item) for item in states}
            if states is not None
            else None
        )
        with self.store.lock:
            return [
                task
                for task in self.store.tasks.values()
                if task.tenant_id == context.tenant_id
                and (state_filter is None or task.state in state_filter)
            ]
    def transition_task(
        self,
        context: RequestContext,
        *,
        task_id: str,
        target_state: TaskState | str,
        expected_revision: int | None = None,
        result: Mapping[str, Any] | None = None,
        error: Mapping[str, Any] | None = None,
    ) -> TaskRecord:
        target = TaskState(target_state)
        action_name = {
            TaskState.APPROVED: "task.approve",
            TaskState.RUNNING: "task.start",
            TaskState.COMPLETED: "task.complete",
            TaskState.FAILED: "task.fail",
            TaskState.CANCELLED: "task.cancel",
        }.get(target)
        if action_name is None:
            raise InvalidTransition(f"cannot transition to {target.value}")
        self._authorize(
            context,
            action_name,
            {"resource_type": "task", "resource_id": task_id},
        )
        with self.store.lock:
            current = self.store.tasks.get(task_id)
            if current is None:
                raise NotFound("task not found")
            self._assert_same_tenant(context, current.tenant_id)
            if expected_revision is not None and current.revision != expected_revision:
                raise RevisionConflict(
                    f"expected revision {expected_revision}, current {current.revision}"
                )
            if target not in ALLOWED_TASK_TRANSITIONS[current.state]:
                raise InvalidTransition(
                    f"invalid transition {current.state.value} -> {target.value}"
                )
            if target == TaskState.COMPLETED and result is None:
                raise ValueError("completed task requires result")
            if target == TaskState.FAILED and error is None:
                raise ValueError("failed task requires error")
            updated = TaskRecord(
                task_id=current.task_id,
                tenant_id=current.tenant_id,
                agent_id=current.agent_id,
                action=current.action,
                payload=current.payload,
                state=target,
                revision=current.revision + 1,
                idempotency_key=current.idempotency_key,
                request_digest=current.request_digest,
                result=copy.deepcopy(dict(result)) if result is not None else current.result,
                error=copy.deepcopy(dict(error)) if error is not None else current.error,
                created_by=current.created_by,
                approved_by=(
                    context.principal_id
                    if target == TaskState.APPROVED
                    else current.approved_by
                ),
                created_at=current.created_at,
                updated_at=_utc_now(),
            )
            self.store.tasks[task_id] = updated
            self._audit(
                context,
                action=action_name,
                resource_type="task",
                resource_id=task_id,
                outcome="allowed",
                details={
                    "from": current.state.value,
                    "to": target.value,
                    "revision": updated.revision,
                },
            )
            return updated
    def list_audit_events(
        self,
        context: RequestContext,
        *,
        resource_id: str | None = None,
    ) -> list[AuditEvent]:
        self._authorize(
            context,
            "audit.read",
            {"resource_type": "audit", "resource_id": resource_id or context.tenant_id},
        )
        with self.store.lock:
            return [
                event
                for event in self.store.audit_events
                if event.tenant_id == context.tenant_id
                and (resource_id is None or event.resource_id == resource_id)
            ]
@@ -0,0 +1,195 @@
 """Git-backed agent execution evidence for AIOS V2.
 Evidence is tenant-scoped, append-friendly, and contains no runtime credentials.
 It provides a deterministic source for claim, heartbeat, progress, result and
 verification state across Codex, OpenClaw and Hermes executors.
 """
 from __future__ import annotations
 import json
 import os
 import threading
 from dataclasses import asdict, dataclass, field, replace
 from datetime import datetime, timezone
 from pathlib import Path
 from typing import Any, Mapping
 def utc_now() -> str:
    return datetime.now(timezone.utc).isoformat().replace("+00:00", "Z")
 def _required(value: str, label: str) -> str:
    normalized = str(value or "").strip()
    if not normalized:
        raise ValueError(f"{label} is required")
    return normalized
 def _safe(value: Mapping[str, Any] | None) -> dict[str, Any]:
    return json.loads(json.dumps(dict(value or {}), ensure_ascii=False, default=str))
@dataclass(frozen=True)
 class ExecutionEvidence:
    execution_id: str
    tenant_id: str
    task_id: str
    agent_id: str
    executor: str
    branch: str
    status: str = "claimed"
    progress_completed: int = 0
    progress_total: int = 1
    current_step: str = "claim"
    last_commit: str | None = None
    test_status: str = "not_run"
    test_summary: Mapping[str, Any] = field(default_factory=dict)
    result_summary: Mapping[str, Any] = field(default_factory=dict)
    claimed_at: str = field(default_factory=utc_now)
    heartbeat_at: str = field(default_factory=utc_now)
    completed_at: str | None = None
    def __post_init__(self) -> None:
        for name in (
            "execution_id",
            "tenant_id",
            "task_id",
            "agent_id",
            "executor",
            "branch",
            "status",
            "current_step",
            "claimed_at",
            "heartbeat_at",
        ):
            object.__setattr__(self, name, _required(getattr(self, name), name))
        if self.progress_completed < 0 or self.progress_total < 1:
            raise ValueError("invalid progress values")
        if self.progress_completed > self.progress_total:
            raise ValueError("progress_completed cannot exceed progress_total")
        object.__setattr__(self, "test_summary", _safe(self.test_summary))
        object.__setattr__(self, "result_summary", _safe(self.result_summary))
 class ExecutionTracker:
    """Persist one JSON evidence record per tenant and execution."""
    def __init__(self, root: str | Path) -> None:
        self.root = Path(root).resolve()
        self.root.mkdir(parents=True, exist_ok=True)
        self._lock = threading.RLock()
    @staticmethod
    def _component(value: str) -> str:
        normalized = _required(value, "path component")
        if normalized in {".", ".."} or "/" in normalized or "\\" in normalized:
            raise ValueError("invalid evidence path component")
        return normalized
    def _path(self, tenant_id: str, execution_id: str) -> Path:
        tenant = self._component(tenant_id)
        execution = self._component(execution_id)
        return self.root / tenant / f"{execution}.json"
    def claim(self, evidence: ExecutionEvidence) -> ExecutionEvidence:
        path = self._path(evidence.tenant_id, evidence.execution_id)
        with self._lock:
            if path.exists():
                current = self._read(path)
                if current.task_id != evidence.task_id or current.agent_id != evidence.agent_id:
                    raise RuntimeError("execution_id already belongs to another task")
                return current
            self._write(path, evidence)
            return evidence
    def heartbeat(
        self,
        *,
        tenant_id: str,
        execution_id: str,
        status: str | None = None,
        current_step: str | None = None,
        progress_completed: int | None = None,
        progress_total: int | None = None,
        last_commit: str | None = None,
    ) -> ExecutionEvidence:
        path = self._path(tenant_id, execution_id)
        with self._lock:
            current = self._read(path)
            updated = replace(
                current,
                status=status or current.status,
                current_step=current_step or current.current_step,
                progress_completed=(
                    progress_completed
                    if progress_completed is not None
                    else current.progress_completed
                ),
                progress_total=(
                    progress_total if progress_total is not None else current.progress_total
                ),
                last_commit=last_commit or current.last_commit,
                heartbeat_at=utc_now(),
            )
            self._write(path, updated)
            return updated
    def complete(
        self,
        *,
        tenant_id: str,
        execution_id: str,
        status: str,
        test_status: str,
        test_summary: Mapping[str, Any] | None = None,
        result_summary: Mapping[str, Any] | None = None,
        last_commit: str | None = None,
    ) -> ExecutionEvidence:
        path = self._path(tenant_id, execution_id)
        with self._lock:
            current = self._read(path)
            updated = replace(
                current,
                status=_required(status, "status"),
                current_step="complete",
                progress_completed=current.progress_total,
                last_commit=last_commit or current.last_commit,
                test_status=_required(test_status, "test_status"),
                test_summary=_safe(test_summary),
                result_summary=_safe(result_summary),
                heartbeat_at=utc_now(),
                completed_at=utc_now(),
            )
            self._write(path, updated)
            return updated
    def get(self, *, tenant_id: str, execution_id: str) -> ExecutionEvidence:
        return self._read(self._path(tenant_id, execution_id))
    def list(self, *, tenant_id: str) -> list[ExecutionEvidence]:
        tenant_dir = self.root / self._component(tenant_id)
        if not tenant_dir.exists():
            return []
        records = [self._read(path) for path in tenant_dir.glob("*.json")]
        return sorted(records, key=lambda item: (item.heartbeat_at, item.execution_id), reverse=True)
    @staticmethod
    def _read(path: Path) -> ExecutionEvidence:
        try:
            value = json.loads(path.read_text(encoding="utf-8"))
        except FileNotFoundError as exc:
            raise KeyError("execution evidence not found") from exc
        if not isinstance(value, Mapping):
            raise RuntimeError("invalid execution evidence file")
        return ExecutionEvidence(**dict(value))
    @staticmethod
    def _write(path: Path, evidence: ExecutionEvidence) -> None:
        path.parent.mkdir(parents=True, exist_ok=True)
        temporary = path.with_suffix(f".tmp-{os.getpid()}")
        temporary.write_text(
            json.dumps(asdict(evidence), ensure_ascii=False, sort_keys=True, indent=2) + "\n",
            encoding="utf-8",
        )
        os.replace(temporary, path)
@@ -0,0 +1,346 @@
 """HTTP/JSON gateway for the AIOS V2 control plane.
 The gateway is intentionally runtime-neutral: browsers and external clients
 call this API, while OpenClaw and Hermes remain behind authenticated adapters.
 Identity headers are accepted only from a trusted reverse proxy in v1; direct
 public exposure is prohibited until signed identity middleware is enabled.
 """
 from __future__ import annotations
 import dataclasses
 import json
 import os
 import re
 import urllib.parse
 from enum import Enum
 from http import HTTPStatus
 from http.server import BaseHTTPRequestHandler, ThreadingHTTPServer
 from typing import Any, Mapping
 from .control_plane import (
    ControlPlaneError,
    ControlPlaneService,
    IdempotencyConflict,
    InvalidTransition,
    NotFound,
    PolicyDenied,
    RequestContext,
    RevisionConflict,
    TaskState,
    TenantIsolationError,
    Unauthorized,
 )
 TENANT_HEADER = "X-AIOS-Tenant-ID"
 PRINCIPAL_HEADER = "X-AIOS-Principal-ID"
 ROLES_HEADER = "X-AIOS-Roles"
 TRACE_HEADER = "X-AIOS-Trace-ID"
 SESSION_HEADER = "X-AIOS-Session-ID"
 POLICY_HEADER = "X-AIOS-Policy-Version"
 AUTHENTICATED_HEADER = "X-AIOS-Authenticated"
 _RESOURCE_ID_RE = re.compile(r"^[A-Za-z0-9][A-Za-z0-9._:-]{0,191}$")
 def _json_value(value: Any) -> Any:
    if dataclasses.is_dataclass(value):
        return {
            field.name: _json_value(getattr(value, field.name))
            for field in dataclasses.fields(value)
        }
    if isinstance(value, Enum):
        return value.value
    if isinstance(value, Mapping):
        return {str(key): _json_value(item) for key, item in value.items()}
    if isinstance(value, (list, tuple, set, frozenset)):
        return [_json_value(item) for item in value]
    return value
 def _require_mapping(value: Any, *, label: str = "JSON body") -> Mapping[str, Any]:
    if not isinstance(value, Mapping):
        raise ValueError(f"{label} must be an object")
    return value
 def _require_resource_id(value: Any, *, label: str) -> str:
    normalized = str(value or "").strip()
    if not _RESOURCE_ID_RE.fullmatch(normalized):
        raise ValueError(f"invalid {label}")
    return normalized
 class GatewayApplication:
    """Pure request router that can be tested without a listening socket."""
    def __init__(self, service: ControlPlaneService | None = None) -> None:
        self.service = service or ControlPlaneService()
    @staticmethod
    def context_from_headers(headers: Mapping[str, str]) -> RequestContext:
        authenticated = str(headers.get(AUTHENTICATED_HEADER, "")).strip().lower()
        if authenticated not in {"1", "true", "yes"}:
            raise Unauthorized("trusted authentication assertion is required")
        roles = [
            role.strip()
            for role in str(headers.get(ROLES_HEADER, "")).split(",")
            if role.strip()
        ]
        return RequestContext.create(
            tenant_id=str(headers.get(TENANT_HEADER, "")),
            principal_id=str(headers.get(PRINCIPAL_HEADER, "")),
            trace_id=str(headers.get(TRACE_HEADER, "")).strip() or None,
            session_id=str(headers.get(SESSION_HEADER, "")).strip() or None,
            policy_version=str(headers.get(POLICY_HEADER, "v1")).strip() or "v1",
            roles=roles,
        )
    def handle(
        self,
        method: str,
        raw_path: str,
        headers: Mapping[str, str],
        body: Any = None,
    ) -> tuple[int, Mapping[str, Any]]:
        parsed = urllib.parse.urlparse(raw_path)
        path = parsed.path.rstrip("/") or "/"
        query = urllib.parse.parse_qs(parsed.query, keep_blank_values=False)
        method = method.upper()
        if method == "GET" and path in {"/healthz", "/api/v2/health"}:
            return HTTPStatus.OK, {
                "service": "aios-v2-control-plane",
                "status": "ok",
                "store": type(self.service.store).__name__,
            }
        context = self.context_from_headers(headers)
        payload = _require_mapping(body or {})
        if method == "POST" and path == "/api/v2/tenants":
            tenant = self.service.register_tenant(
                context,
                tenant_id=_require_resource_id(payload.get("tenant_id"), label="tenant_id"),
                display_name=str(payload.get("display_name", "")),
                metadata=_require_mapping(payload.get("metadata", {}), label="metadata"),
            )
            return HTTPStatus.CREATED, {"tenant": _json_value(tenant)}
        if method == "GET" and path == "/api/v2/tenants/current":
            tenant = self.service.get_tenant(context)
            return HTTPStatus.OK, {"tenant": _json_value(tenant)}
        if method == "POST" and path == "/api/v2/agents":
            agent = self.service.register_agent(
                context,
                agent_id=_require_resource_id(payload.get("agent_id"), label="agent_id"),
                display_name=str(payload.get("display_name", "")),
                runtime=str(payload.get("runtime", "")),
                capabilities=payload.get("capabilities", ()),
                metadata=_require_mapping(payload.get("metadata", {}), label="metadata"),
            )
            return HTTPStatus.CREATED, {"agent": _json_value(agent)}
        if method == "GET" and path == "/api/v2/agents":
            return HTTPStatus.OK, {
                "agents": _json_value(self.service.list_agents(context))
            }
        agent_match = re.fullmatch(r"/api/v2/agents/([^/]+)", path)
        if method == "GET" and agent_match:
            agent_id = _require_resource_id(
                urllib.parse.unquote(agent_match.group(1)), label="agent_id"
            )
            return HTTPStatus.OK, {
                "agent": _json_value(self.service.get_agent(context, agent_id))
            }
        if method == "POST" and path == "/api/v2/tasks":
            task = self.service.create_task(
                context,
                agent_id=_require_resource_id(payload.get("agent_id"), label="agent_id"),
                action=str(payload.get("action", "")).strip(),
                payload=_require_mapping(payload.get("payload", {}), label="payload"),
                idempotency_key=(
                    str(payload["idempotency_key"]).strip()
                    if payload.get("idempotency_key") is not None
                    else None
                ),
            )
            return HTTPStatus.CREATED, {"task": _json_value(task)}
        if method == "GET" and path == "/api/v2/tasks":
            requested_states = query.get("state")
            tasks = self.service.list_tasks(context, states=requested_states)
            return HTTPStatus.OK, {"tasks": _json_value(tasks)}
        task_match = re.fullmatch(r"/api/v2/tasks/([^/]+)", path)
        if method == "GET" and task_match:
            task_id = _require_resource_id(
                urllib.parse.unquote(task_match.group(1)), label="task_id"
            )
            return HTTPStatus.OK, {
                "task": _json_value(self.service.get_task(context, task_id))
            }
        transition_match = re.fullmatch(
            r"/api/v2/tasks/([^/]+)/(approve|start|complete|fail|cancel)", path
        )
        if method == "POST" and transition_match:
            task_id = _require_resource_id(
                urllib.parse.unquote(transition_match.group(1)), label="task_id"
            )
            transition = transition_match.group(2)
            target = {
                "approve": TaskState.APPROVED,
                "start": TaskState.RUNNING,
                "complete": TaskState.COMPLETED,
                "fail": TaskState.FAILED,
                "cancel": TaskState.CANCELLED,
            }[transition]
            expected_revision = payload.get("expected_revision")
            if expected_revision is not None:
                expected_revision = int(expected_revision)
            task = self.service.transition_task(
                context,
                task_id=task_id,
                target_state=target,
                expected_revision=expected_revision,
                result=(
                    _require_mapping(payload.get("result"), label="result")
                    if payload.get("result") is not None
                    else None
                ),
                error=(
                    _require_mapping(payload.get("error"), label="error")
                    if payload.get("error") is not None
                    else None
                ),
            )
            return HTTPStatus.OK, {"task": _json_value(task)}
        if method == "GET" and path == "/api/v2/audit":
            resource_id = query.get("resource_id", [None])[0]
            events = self.service.list_audit_events(
                context,
                resource_id=(
                    _require_resource_id(resource_id, label="resource_id")
                    if resource_id
                    else None
                ),
            )
            return HTTPStatus.OK, {"events": _json_value(events)}
        return HTTPStatus.NOT_FOUND, {
            "error": "not_found",
            "message": "route not found",
        }
 _ERROR_STATUS: tuple[tuple[type[BaseException], HTTPStatus], ...] = (
    (Unauthorized, HTTPStatus.UNAUTHORIZED),
    (PolicyDenied, HTTPStatus.FORBIDDEN),
    (TenantIsolationError, HTTPStatus.NOT_FOUND),
    (NotFound, HTTPStatus.NOT_FOUND),
    (IdempotencyConflict, HTTPStatus.CONFLICT),
    (RevisionConflict, HTTPStatus.CONFLICT),
    (InvalidTransition, HTTPStatus.CONFLICT),
    (ValueError, HTTPStatus.BAD_REQUEST),
    (ControlPlaneError, HTTPStatus.BAD_REQUEST),
 )
 def error_response(exc: BaseException) -> tuple[int, Mapping[str, Any]]:
    for error_type, status in _ERROR_STATUS:
        if isinstance(exc, error_type):
            return status, {
                "error": error_type.__name__,
                "message": str(exc),
            }
    return HTTPStatus.INTERNAL_SERVER_ERROR, {
        "error": "InternalServerError",
        "message": "internal server error",
    }
 class GatewayRequestHandler(BaseHTTPRequestHandler):
    server_version = "AIOSV2Gateway/1.0"
    max_body_bytes = 1_048_576
    @property
    def application(self) -> GatewayApplication:
        return self.server.application  # type: ignore[attr-defined]
    def _read_body(self) -> Any:
        length = int(self.headers.get("Content-Length", "0") or "0")
        if length < 0 or length > self.max_body_bytes:
            raise ValueError("request body is too large")
        if length == 0:
            return {}
        content_type = self.headers.get("Content-Type", "").split(";", 1)[0].strip()
        if content_type != "application/json":
            raise ValueError("Content-Type must be application/json")
        raw = self.rfile.read(length)
        return json.loads(raw.decode("utf-8"))
    def _send_json(self, status: int, payload: Mapping[str, Any]) -> None:
        encoded = json.dumps(
            _json_value(payload), ensure_ascii=False, separators=(",", ":")
        ).encode("utf-8")
        self.send_response(int(status))
        self.send_header("Content-Type", "application/json; charset=utf-8")
        self.send_header("Content-Length", str(len(encoded)))
        self.send_header("Cache-Control", "no-store")
        self.send_header("X-Content-Type-Options", "nosniff")
        self.end_headers()
        self.wfile.write(encoded)
    def _dispatch(self, method: str) -> None:
        try:
            body = self._read_body() if method in {"POST", "PUT", "PATCH"} else {}
            status, payload = self.application.handle(
                method,
                self.path,
                {key: value for key, value in self.headers.items()},
                body,
            )
        except BaseException as exc:  # handler boundary; sanitized below
            status, payload = error_response(exc)
        self._send_json(status, payload)
    def do_GET(self) -> None:  # noqa: N802
        self._dispatch("GET")
    def do_POST(self) -> None:  # noqa: N802
        self._dispatch("POST")
    def log_message(self, format: str, *args: Any) -> None:
        # Production deployment routes structured access logs through the
        # reverse proxy; do not log identity headers or request bodies here.
        super().log_message(format, *args)
 class GatewayHTTPServer(ThreadingHTTPServer):
    daemon_threads = True
    allow_reuse_address = True
    def __init__(
        self,
        server_address: tuple[str, int],
        application: GatewayApplication,
    ) -> None:
        self.application = application
        super().__init__(server_address, GatewayRequestHandler)
 def main() -> None:
    host = os.getenv("AIOS_V2_GATEWAY_HOST", "127.0.0.1")
    port = int(os.getenv("AIOS_V2_GATEWAY_PORT", "18920"))
    server = GatewayHTTPServer((host, port), GatewayApplication())
    print(f"AIOS V2 control-plane gateway listening on {host}:{port}")
    server.serve_forever()
 if __name__ == "__main__":
    main()
@@ -0,0 +1,123 @@
 """Sanitized GitOps runtime status for the AIOS Web control plane."""
 from __future__ import annotations
 import os
 import subprocess
 from dataclasses import dataclass, asdict
 from pathlib import Path
 from typing import Any, Mapping
@dataclass(frozen=True)
 class GitOpsStatus:
    available: bool
    service_status: str
    workspace_mode: str
    workspace_branch: str | None
    workspace_ref: str | None
    workspace_dirty: bool | None
    candidate_branch: str
    release_channel: str
    release_version: str
    sync_status: str
    details: Mapping[str, Any]
    def as_dict(self) -> dict[str, Any]:
        return asdict(self)
 def _run_git(worktree: Path, *args: str) -> tuple[int, str, str]:
    process = subprocess.run(
        ["git", "-C", str(worktree), *args],
        text=True,
        capture_output=True,
        timeout=5,
    )
    return process.returncode, process.stdout.strip(), process.stderr.strip()
 def collect_gitops_status() -> GitOpsStatus:
    """Return only non-sensitive Git/workspace metadata.
    Remote URLs, environment values, credentials, filesystem contents and
    command stderr are deliberately excluded from the Web response.
    """
    candidate_branch = os.getenv(
        "AIOS_CANDIDATE_BRANCH",
        "integration/aios-v2r1-cto-production-candidate-20260720",
    )
    release_channel = os.getenv("AIOS_RELEASE_CHANNEL", "candidate")
    release_version = os.getenv("AIOS_RELEASE_VERSION", "V2R1C0-candidate")
    sync_status = os.getenv("AIOS_GITOPS_SYNC_STATUS", "unknown")
    worktree = Path(
        os.getenv(
            "AIOS_GITOPS_CONTROLLER_WORKTREE",
            "/data/aios-gitops/worktrees/controller",
        )
    )
    if not worktree.is_dir():
        return GitOpsStatus(
            available=False,
            service_status="unknown",
            workspace_mode="unavailable",
            workspace_branch=None,
            workspace_ref=None,
            workspace_dirty=None,
            candidate_branch=candidate_branch,
            release_channel=release_channel,
            release_version=release_version,
            sync_status=sync_status,
            details={"reason": "controller_worktree_unavailable"},
        )
    inside_code, inside, _ = _run_git(worktree, "rev-parse", "--is-inside-work-tree")
    if inside_code != 0 or inside != "true":
        return GitOpsStatus(
            available=False,
            service_status="unknown",
            workspace_mode="invalid",
            workspace_branch=None,
            workspace_ref=None,
            workspace_dirty=None,
            candidate_branch=candidate_branch,
            release_channel=release_channel,
            release_version=release_version,
            sync_status=sync_status,
            details={"reason": "controller_worktree_invalid"},
        )
    branch_code, branch, _ = _run_git(worktree, "branch", "--show-current")
    ref_code, ref_value, _ = _run_git(worktree, "rev-parse", "HEAD")
    dirty_code, dirty_value, _ = _run_git(worktree, "status", "--porcelain=v1")
    service_status = "unknown"
    try:
        process = subprocess.run(
            ["systemctl", "is-active", "aios-gitops-model-api"],
            text=True,
            capture_output=True,
            timeout=3,
        )
        service_status = process.stdout.strip() or "inactive"
    except (FileNotFoundError, subprocess.TimeoutExpired):
        service_status = "unavailable"
    branch_value = branch if branch_code == 0 and branch else None
    return GitOpsStatus(
        available=True,
        service_status=service_status,
        workspace_mode="branch" if branch_value else "detached",
        workspace_branch=branch_value,
        workspace_ref=ref_value if ref_code == 0 else None,
        workspace_dirty=bool(dirty_value) if dirty_code == 0 else None,
        candidate_branch=candidate_branch,
        release_channel=release_channel,
        release_version=release_version,
        sync_status=sync_status,
        details={
            "controller_clean": dirty_code == 0 and not dirty_value,
            "controller_detached": branch_code == 0 and not branch,
        },
    )
@@ -0,0 +1,264 @@
 """AIOS V2 execution orchestrator.
 The orchestrator is the only component allowed to translate an approved AIOS
 task into an OpenClaw command or Hermes job. OpenClaw and Hermes remain pure
 execution runtimes and never own tenant policy or task lifecycle.
 """
 from __future__ import annotations
 from dataclasses import dataclass
 from typing import Any, Mapping
 from aios_hermes import HermesAdapter, HermesJob, HermesResult
 from aios_openclaw import (
    CommandState,
    OpenClawAdapter,
    OpenClawCommand,
    OpenClawResult,
    sign_command,
 )
 from .control_plane import (
    AgentRecord,
    ControlPlaneService,
    RequestContext,
    TaskRecord,
    TaskState,
 )
 class OrchestrationError(RuntimeError):
    """AIOS failed to dispatch or reconcile a runtime execution."""
@dataclass(frozen=True)
 class DispatchRecord:
    tenant_id: str
    task_id: str
    agent_id: str
    runtime: str
    runtime_execution_id: str
    state: str
    trace_id: str
 class ExecutionOrchestrator:
    """Dispatch approved AIOS tasks to runtime adapters and reconcile results."""
    def __init__(
        self,
        control_plane: ControlPlaneService,
        *,
        openclaw: OpenClawAdapter | None = None,
        hermes: HermesAdapter | None = None,
        openclaw_signing_material_resolver=None,
        openclaw_key_id: str = "default",
        default_cell_id: str = "default",
        default_profile_id: str = "default",
    ) -> None:
        self.control_plane = control_plane
        self.openclaw = openclaw
        self.hermes = hermes
        self.openclaw_signing_material_resolver = openclaw_signing_material_resolver
        self.openclaw_key_id = openclaw_key_id
        self.default_cell_id = default_cell_id
        self.default_profile_id = default_profile_id
        self._dispatches: dict[tuple[str, str], DispatchRecord] = {}
    def dispatch(
        self,
        context: RequestContext,
        *,
        task_id: str,
        timeout_seconds: float | None = None,
    ) -> DispatchRecord:
        task = self.control_plane.get_task(context, task_id)
        if task.state != TaskState.APPROVED:
            raise OrchestrationError("only APPROVED tasks may be dispatched")
        agent = self.control_plane.get_agent(context, task.agent_id)
        existing = self._dispatches.get((task.tenant_id, task.task_id))
        if existing is not None:
            return existing
        running = self.control_plane.transition_task(
            context,
            task_id=task.task_id,
            target_state=TaskState.RUNNING,
            expected_revision=task.revision,
        )
        runtime = agent.runtime.strip().lower()
        try:
            if runtime == "openclaw":
                record = self._dispatch_openclaw(
                    context,
                    agent,
                    running,
                    timeout_seconds=timeout_seconds,
                )
            elif runtime == "hermes":
                record = self._dispatch_hermes(context, agent, running)
            else:
                raise OrchestrationError(f"unsupported runtime: {agent.runtime}")
        except Exception as exc:
            current = self.control_plane.get_task(context, task.task_id)
            if current.state == TaskState.RUNNING:
                self.control_plane.transition_task(
                    context,
                    task_id=task.task_id,
                    target_state=TaskState.FAILED,
                    expected_revision=current.revision,
                    error={"code": "dispatch_failed", "message": str(exc)},
                )
            raise
        self._dispatches[(task.tenant_id, task.task_id)] = record
        return record
    def _dispatch_openclaw(
        self,
        context: RequestContext,
        agent: AgentRecord,
        task: TaskRecord,
        *,
        timeout_seconds: float | None,
    ) -> DispatchRecord:
        if self.openclaw is None or self.openclaw_signing_material_resolver is None:
            raise OrchestrationError("OpenClaw runtime is not configured")
        key_material = self.openclaw_signing_material_resolver(
            task.tenant_id,
            self.openclaw_key_id,
        )
        command = OpenClawCommand.create(
            tenant_id=task.tenant_id,
            task_id=task.task_id,
            trace_id=context.trace_id,
            action=task.action,
            payload=task.payload,
            idempotency_key=task.idempotency_key or task.task_id,
            key_id=self.openclaw_key_id,
        )
        command = sign_command(command, key_material)
        result = self.openclaw.execute(command, timeout_seconds=timeout_seconds)
        self.reconcile_openclaw(context, result)
        return DispatchRecord(
            tenant_id=task.tenant_id,
            task_id=task.task_id,
            agent_id=agent.agent_id,
            runtime="openclaw",
            runtime_execution_id=command.command_id,
            state=result.state.value,
            trace_id=context.trace_id,
        )
    def _dispatch_hermes(
        self,
        context: RequestContext,
        agent: AgentRecord,
        task: TaskRecord,
    ) -> DispatchRecord:
        if self.hermes is None:
            raise OrchestrationError("Hermes runtime is not configured")
        metadata = dict(agent.metadata)
        job = HermesJob.create(
            tenant_id=task.tenant_id,
            cell_id=str(metadata.get("cell_id") or self.default_cell_id),
            profile_id=str(metadata.get("profile_id") or self.default_profile_id),
            task_id=task.task_id,
            trace_id=context.trace_id,
            action=task.action,
            payload=task.payload,
            idempotency_key=task.idempotency_key or task.task_id,
            priority=int(metadata.get("priority", 100)),
        )
        queued = self.hermes.enqueue(job)
        return DispatchRecord(
            tenant_id=task.tenant_id,
            task_id=task.task_id,
            agent_id=agent.agent_id,
            runtime="hermes",
            runtime_execution_id=queued.job_id,
            state=queued.state.value,
            trace_id=context.trace_id,
        )
    def reconcile_openclaw(
        self,
        context: RequestContext,
        result: OpenClawResult,
    ) -> TaskRecord:
        task = self.control_plane.get_task(context, result.task_id)
        if result.tenant_id != task.tenant_id:
            raise OrchestrationError("OpenClaw result tenant mismatch")
        if task.state != TaskState.RUNNING:
            return task
        if result.state == CommandState.COMPLETED:
            return self.control_plane.transition_task(
                context,
                task_id=task.task_id,
                target_state=TaskState.COMPLETED,
                expected_revision=task.revision,
                result=result.output or {},
            )
        if result.state in {
            CommandState.FAILED,
            CommandState.TIMED_OUT,
            CommandState.REJECTED,
        }:
            return self.control_plane.transition_task(
                context,
                task_id=task.task_id,
                target_state=TaskState.FAILED,
                expected_revision=task.revision,
                error=result.error or {"code": result.state.value.lower()},
            )
        if result.state == CommandState.CANCELLED:
            return self.control_plane.transition_task(
                context,
                task_id=task.task_id,
                target_state=TaskState.CANCELLED,
                expected_revision=task.revision,
            )
        return task
    def reconcile_hermes(
        self,
        context: RequestContext,
        result: HermesResult,
    ) -> TaskRecord:
        task = self.control_plane.get_task(context, result.task_id)
        if result.tenant_id != task.tenant_id:
            raise OrchestrationError("Hermes result tenant mismatch")
        if task.state != TaskState.RUNNING:
            return task
        if result.state.value == "COMPLETED":
            return self.control_plane.transition_task(
                context,
                task_id=task.task_id,
                target_state=TaskState.COMPLETED,
                expected_revision=task.revision,
                result=result.output or {},
            )
        if result.state.value == "FAILED":
            return self.control_plane.transition_task(
                context,
                task_id=task.task_id,
                target_state=TaskState.FAILED,
                expected_revision=task.revision,
                error=result.error or {"code": "hermes_failed"},
            )
        if result.state.value == "CANCELLED":
            return self.control_plane.transition_task(
                context,
                task_id=task.task_id,
                target_state=TaskState.CANCELLED,
                expected_revision=task.revision,
            )
        return task
    def dispatch_snapshot(self, *, tenant_id: str) -> Mapping[str, DispatchRecord]:
        return {
            task_id: record
            for (record_tenant, task_id), record in self._dispatches.items()
            if record_tenant == tenant_id
        }
@@ -0,0 +1,181 @@
 """Durable single-process persistence for the AIOS V2 control plane.
 The candidate uses an atomic SQLite snapshot so a gateway restart does not lose
 tenants, agents, task state, idempotency records, or audit events.  The domain
 service remains storage-agnostic; a later HA backend can replace this module
 without changing API or runtime-adapter contracts.
 """
 from __future__ import annotations
 import json
 import sqlite3
 import threading
 from dataclasses import asdict
 from pathlib import Path
 from typing import Any, Callable
 from .control_plane import (
    AgentRecord,
    AuditEvent,
    ControlPlaneService,
    InMemoryControlPlaneStore,
    TaskRecord,
    TaskState,
    TenantRecord,
 )
 class SQLiteSnapshotRepository:
    """Store one versioned control-plane snapshot in SQLite."""
    SCHEMA_VERSION = 1
    def __init__(self, path: str | Path) -> None:
        self.path = Path(path).resolve()
        self.path.parent.mkdir(parents=True, exist_ok=True)
        self._lock = threading.RLock()
        self._initialize()
    def _connect(self) -> sqlite3.Connection:
        connection = sqlite3.connect(self.path, timeout=30.0)
        connection.execute("PRAGMA journal_mode=WAL")
        connection.execute("PRAGMA synchronous=FULL")
        connection.execute("PRAGMA foreign_keys=ON")
        return connection
    def _initialize(self) -> None:
        with self._connect() as connection:
            connection.execute(
                """
                CREATE TABLE IF NOT EXISTS control_plane_snapshot (
                    singleton_id INTEGER PRIMARY KEY CHECK (singleton_id = 1),
                    schema_version INTEGER NOT NULL,
                    payload_json TEXT NOT NULL,
                    updated_at TEXT NOT NULL DEFAULT CURRENT_TIMESTAMP
                )
                """
            )
    def load(self) -> dict[str, Any] | None:
        with self._lock, self._connect() as connection:
            row = connection.execute(
                "SELECT schema_version, payload_json FROM control_plane_snapshot WHERE singleton_id = 1"
            ).fetchone()
        if row is None:
            return None
        schema_version, payload_json = row
        if int(schema_version) != self.SCHEMA_VERSION:
            raise RuntimeError(
                f"unsupported control-plane snapshot version: {schema_version}"
            )
        decoded = json.loads(str(payload_json))
        if not isinstance(decoded, dict):
            raise RuntimeError("invalid control-plane snapshot")
        return decoded
    def save(self, payload: dict[str, Any]) -> None:
        encoded = json.dumps(
            payload,
            ensure_ascii=False,
            sort_keys=True,
            separators=(",", ":"),
            default=str,
        )
        with self._lock, self._connect() as connection:
            connection.execute("BEGIN IMMEDIATE")
            connection.execute(
                """
                INSERT INTO control_plane_snapshot (
                    singleton_id, schema_version, payload_json, updated_at
                ) VALUES (1, ?, ?, CURRENT_TIMESTAMP)
                ON CONFLICT(singleton_id) DO UPDATE SET
                    schema_version = excluded.schema_version,
                    payload_json = excluded.payload_json,
                    updated_at = CURRENT_TIMESTAMP
                """,
                (self.SCHEMA_VERSION, encoded),
            )
            connection.commit()
 class DurableControlPlaneService(ControlPlaneService):
    """Control-plane service that persists after every successful mutation."""
    def __init__(
        self,
        repository: SQLiteSnapshotRepository,
        *,
        policy_evaluator=None,
    ) -> None:
        self.repository = repository
        store = InMemoryControlPlaneStore()
        self._restore(store, repository.load())
        super().__init__(store, policy_evaluator=policy_evaluator)
    @staticmethod
    def _restore(
        store: InMemoryControlPlaneStore,
        snapshot: dict[str, Any] | None,
    ) -> None:
        if not snapshot:
            return
        for item in snapshot.get("tenants", []):
            record = TenantRecord(**item)
            store.tenants[record.tenant_id] = record
        for item in snapshot.get("agents", []):
            normalized = dict(item)
            normalized["capabilities"] = tuple(normalized.get("capabilities", ()))
            record = AgentRecord(**normalized)
            store.agents[record.agent_id] = record
        for item in snapshot.get("tasks", []):
            normalized = dict(item)
            normalized["state"] = TaskState(normalized["state"])
            record = TaskRecord(**normalized)
            store.tasks[record.task_id] = record
        store.audit_events.extend(AuditEvent(**item) for item in snapshot.get("audit", []))
        for item in snapshot.get("idempotency", []):
            store.idempotency[(item["tenant_id"], item["key"])] = (
                item["task_id"],
                item["digest"],
            )
    def _snapshot(self) -> dict[str, Any]:
        with self.store.lock:
            return {
                "schema_version": self.repository.SCHEMA_VERSION,
                "tenants": [asdict(item) for item in self.store.tenants.values()],
                "agents": [asdict(item) for item in self.store.agents.values()],
                "tasks": [
                    {**asdict(item), "state": item.state.value}
                    for item in self.store.tasks.values()
                ],
                "audit": [asdict(item) for item in self.store.audit_events],
                "idempotency": [
                    {
                        "tenant_id": tenant_id,
                        "key": key,
                        "task_id": task_id,
                        "digest": digest,
                    }
                    for (tenant_id, key), (task_id, digest)
                    in self.store.idempotency.items()
                ],
            }
    def _mutate(self, operation: Callable[[], Any]) -> Any:
        with self.store.lock:
            result = operation()
            self.repository.save(self._snapshot())
            return result
    def register_tenant(self, *args, **kwargs):
        return self._mutate(lambda: super(DurableControlPlaneService, self).register_tenant(*args, **kwargs))
    def register_agent(self, *args, **kwargs):
        return self._mutate(lambda: super(DurableControlPlaneService, self).register_agent(*args, **kwargs))
    def create_task(self, *args, **kwargs):
        return self._mutate(lambda: super(DurableControlPlaneService, self).create_task(*args, **kwargs))
    def transition_task(self, *args, **kwargs):
        return self._mutate(lambda: super(DurableControlPlaneService, self).transition_task(*args, **kwargs))
@@ -1,141 +0,0 @@
 #!/usr/bin/env python3
 """Prove two real central-decision to Gateway-dispatch to Codex lifecycle cycles."""
 from __future__ import annotations
 import argparse
 import json
 from pathlib import Path
 import sqlite3
 import time
 import uuid
 from health_probe_common import (
    ProbeError, gateway_base, load_config, read_token, request_json, state_version,
 )
 REQUIRED_MARKERS = {"CODEX_AUTO_CLAIM", "TASK_STARTED", "TASK_EXECUTED", "TASK_ENDED", "EVENT_POSTED"}
 def marker_counts(path: Path, command_id: str) -> dict[str, int]:
    counts = {marker: 0 for marker in REQUIRED_MARKERS}
    with path.open("r", encoding="utf-8") as handle:
        for line in handle:
            try:
                record = json.loads(line)
            except json.JSONDecodeError:
                continue
            marker = record.get("marker")
            if record.get("command_id") == command_id and marker in counts:
                counts[marker] += 1
    return counts
 def watcher_status(database: Path, command_id: str) -> str | None:
    with sqlite3.connect(f"file:{database}?mode=ro", uri=True, timeout=2) as db:
        row = db.execute("SELECT status FROM commands WHERE command_id=?", (command_id,)).fetchone()
    return None if row is None else str(row[0])
 def wait_for_cycle(config: dict[str, object], token: str, cycle: int) -> tuple[str, str]:
    endpoint = str(config["gateway_event_url"])
    project = str(config["project_key"])
    event_id = f"a0-cycle-{cycle}-" + uuid.uuid4().hex
    work_unit_id = f"A0-CODEX-CYCLE-{cycle}-" + uuid.uuid4().hex[:16]
    event = {
        "schema_version": 1,
        "project_key": project,
        "contract_commit": config["contract_commit"],
        "event_id": event_id,
        "source_agent": "system",
        "event_type": "a0_unattended_acceptance_requested",
        "work_unit_id": work_unit_id,
        "status": "READY",
        "state_version": state_version(endpoint, project, token),
        "implementation_commit": None,
        "result_commit": None,
        "blocker": None,
        "evidence_refs": [],
        "payload": {
            "acceptance_contract": {
                "verdict": "DISPATCH",
                "work_unit_id": work_unit_id,
                "track_id": "A0_AUTOMATION",
                "gate_id": "A0_AUTOMATION_CLOSED",
                "primary_agent": "codex",
                "fallback_agent": "openclaw",
                "clean_base_commit": config["probe_clean_base_commit"],
                "allowed_paths": [],
                "required_checks": ["acceptance_probe"],
                "forbidden_operations": ["production_mutation", "git_mutation"],
                "instruction": "Run the fixed side-effect-free acceptance probe only.",
            }
        },
    }
    request_json(endpoint, token, method="POST", payload=event)
    deadline = time.monotonic() + int(config["cycle_timeout_seconds"])
    channel = None
    while time.monotonic() < deadline:
        _, status = request_json(endpoint + "/" + event_id, token)
        if status.get("status") == "TRIGGER_FAILED":
            raise ProbeError("central_trigger_failed")
        if status.get("status") == "DECIDED" and status.get("decision_applied") is True:
            channel = status.get("trigger_channel")
            break
        time.sleep(float(config["poll_interval_seconds"]))
    if channel not in {"workspace_agent", "responses_api"}:
        raise ProbeError("central_decision_timeout")
    database = Path(str(config["codex_database_path"]))
    audit = Path(str(config["codex_audit_log"]))
    while time.monotonic() < deadline:
        if database.exists() and audit.exists() and watcher_status(database, work_unit_id) == "COMPLETED":
            counts = marker_counts(audit, work_unit_id)
            if all(counts[marker] == 1 for marker in REQUIRED_MARKERS):
                for suffix in ("started", "ended"):
                    _, lifecycle = request_json(endpoint + "/" + work_unit_id + ":" + suffix, token)
                    if lifecycle.get("event_id") != work_unit_id + ":" + suffix:
                        raise ProbeError("gateway_lifecycle_missing")
                return event_id, str(channel)
        time.sleep(float(config["poll_interval_seconds"]))
    raise ProbeError("gateway_dispatch_or_codex_lifecycle_timeout")
 def project_summaries_since(database: Path, project: str, since: int) -> int:
    with sqlite3.connect(f"file:{database}?mode=ro", uri=True, timeout=2) as db:
        row = db.execute(
            "SELECT COUNT(*) FROM lifecycle WHERE project_key=? AND kind='PROJECT_SUMMARY' AND created_at>=?",
            (project, since),
        ).fetchone()
    return 0 if row is None else int(row[0])
 def run(config_path: Path) -> int:
    config = load_config(config_path)
    endpoint = str(config["gateway_event_url"])
    gateway_base(endpoint)
    token = read_token(Path(str(config["gateway_token_file"])))
    started = int(time.time())
    results = [wait_for_cycle(config, token, cycle) for cycle in (1, 2)]
    if len({item[0] for item in results}) != 2:
        raise ProbeError("duplicate_cycle_event")
    if project_summaries_since(
        Path(str(config["gateway_database_path"])), str(config["project_key"]), started
    ) < 2:
        raise ProbeError("project_summary_not_persisted")
    return 0
 def main() -> int:
    parser = argparse.ArgumentParser()
    parser.add_argument("--config", required=True, type=Path)
    args = parser.parse_args()
    try:
        return run(args.config)
    except (KeyError, OSError, ProbeError, sqlite3.Error, ValueError):
        return 1
 if __name__ == "__main__":
    raise SystemExit(main())
@@ -1,171 +0,0 @@
 #!/usr/bin/env python3
 """Prove a real central-model decision through the authenticated Gateway."""
 from __future__ import annotations
 import argparse
 import json
 import os
 import stat
 import sys
 import time
 from pathlib import Path
 import uuid
 from health_probe_common import (
    ProbeError, atomic_receipt, gateway_base, load_config, read_token,
    request_json, state_version,
 )
 PENDING_PATH = Path("/run/aios-health/central-gpt.pending.json")
 def _load_pending(
    path: Path, contract: str, *, owners: set[int] | None = None
 ) -> tuple[str, int] | None:
    allowed_owners = {0} if owners is None else owners
    if not path.exists():
        return None
    info = path.lstat()
    if (
        not path.is_absolute()
        or stat.S_ISLNK(info.st_mode)
        or not stat.S_ISREG(info.st_mode)
        or stat.S_IMODE(info.st_mode) & ~0o640
        or (os.name == "posix" and info.st_uid not in allowed_owners)
        or path.resolve(strict=True) != Path(os.path.abspath(path))
    ):
        raise ProbeError("unsafe_pending_marker")
    value = json.loads(path.read_text(encoding="utf-8"))
    if not isinstance(value, dict) or set(value) != {"contract_commit", "event_id", "state_version"}:
        raise ProbeError("invalid_pending_marker")
    event_id = value["event_id"]
    version = value["state_version"]
    if (
        value["contract_commit"] != contract
        or not isinstance(event_id, str)
        or not event_id.startswith("health-central-")
        or len(event_id) != 47
        or not isinstance(version, int)
        or version < 0
    ):
        raise ProbeError("invalid_pending_marker")
    return event_id, version
 def _clear_pending(path: Path) -> None:
    if not path.exists():
        return
    info = path.lstat()
    if stat.S_ISLNK(info.st_mode) or not stat.S_ISREG(info.st_mode):
        raise ProbeError("unsafe_pending_marker")
    path.unlink()
 def _safe_reason(exc: BaseException) -> str:
    reason = str(exc)
    if reason and len(reason) <= 64 and all(char.islower() or char.isdigit() or char == "_" for char in reason):
        return reason
    return "probe_error"
 def run(config_path: Path, pending_path: Path = PENDING_PATH) -> int:
    config = load_config(config_path)
    endpoint = config["gateway_event_url"]
    gateway_base(endpoint)
    token = read_token(Path(config["gateway_token_file"]))
    project = config["project_key"]
    contract = config["contract_commit"]
    pending = _load_pending(pending_path, contract)
    is_new = pending is None
    if is_new:
        event_id = "health-central-" + uuid.uuid4().hex
        version = state_version(endpoint, project, token)
        atomic_receipt(
            pending_path,
            {"contract_commit": contract, "event_id": event_id, "state_version": version},
        )
    else:
        event_id, version = pending
    event = {
        "schema_version": 1,
        "project_key": project,
        "contract_commit": contract,
        "event_id": event_id,
        "source_agent": "system",
        "event_type": "central_gpt_health_probe",
        "work_unit_id": "CENTRAL-GPT-HEALTH",
        "status": "READY",
        "state_version": version,
        "implementation_commit": None,
        "result_commit": None,
        "blocker": None,
        "evidence_refs": [],
        "payload": {"probe": "side_effect_free", "nonce": event_id},
    }
    should_post = is_new
    if not is_new:
        try:
            _, observed = request_json(endpoint + "/" + event_id, token)
        except ProbeError as exc:
            if str(exc) != "gateway_http_404":
                raise
            should_post = True
    else:
        observed = None
    if should_post:
        try:
            status, _ = request_json(endpoint, token, method="POST", payload=event)
        except ProbeError as exc:
            if str(exc) == "gateway_fail_closed_409":
                _clear_pending(pending_path)
            raise
        if status not in {200, 202}:
            raise ProbeError("event_not_accepted")
        observed = None
    deadline = time.monotonic() + int(config["central_timeout_seconds"])
    while time.monotonic() < deadline:
        _, observed = request_json(endpoint + "/" + event_id, token)
        if observed.get("status") in {"DECIDED", "TRIGGER_FAILED", "DEAD_LETTER"}:
            break
        time.sleep(float(config["poll_interval_seconds"]))
    if isinstance(observed, dict) and observed.get("status") == "DEAD_LETTER":
        _clear_pending(pending_path)
        raise ProbeError("central_event_dead_letter")
    if not isinstance(observed, dict) or observed.get("status") != "DECIDED" or observed.get("decision_applied") is not True:
        raise ProbeError("central_decision_not_proven")
    channel = observed.get("trigger_channel")
    if channel not in {"workspace_agent", "responses_api"}:
        raise ProbeError("central_channel_not_proven")
    _clear_pending(pending_path)
    now = int(time.time())
    atomic_receipt(
        Path(config["central_receipt_path"]),
        {
            "subject": "central_gpt",
            "health": "HEALTHY",
            "live_probe_succeeded": True,
            "observed_at_epoch": now,
            "event_id": event_id,
            "decision_applied": True,
            "state_version_after": observed["state_version_after"],
            "workspace_agent": {"validated_live": channel == "workspace_agent"},
            "responses_api": {"validated_live": channel == "responses_api"},
        },
    )
    return 0
 def main() -> int:
    parser = argparse.ArgumentParser()
    parser.add_argument("--config", required=True, type=Path)
    args = parser.parse_args()
    try:
        return run(args.config)
    except (KeyError, OSError, ProbeError, ValueError) as exc:
        print(f"CENTRAL_GPT_PROBE=BLOCKED reason={_safe_reason(exc)}", file=sys.stderr)
        return 1
 if __name__ == "__main__":
    raise SystemExit(main())
@@ -1,113 +0,0 @@
 #!/usr/bin/env python3
 """Queue one fixed no-side-effect command and prove the watcher lifecycle."""
 from __future__ import annotations
 import argparse
 import json
 import os
 from pathlib import Path
 import sqlite3
 import stat
 import time
 import uuid
 from health_probe_common import (
    ProbeError, atomic_receipt, gateway_base, load_config, read_token,
    request_json, state_version,
 )
 MARKERS = {"CODEX_AUTO_CLAIM", "TASK_STARTED", "TASK_EXECUTED", "TASK_ENDED", "EVENT_POSTED"}
 def _write_command(path: Path, value: dict[str, object]) -> None:
    directory = path.parent
    info = directory.lstat()
    if stat.S_ISLNK(info.st_mode) or not stat.S_ISDIR(info.st_mode) or stat.S_IMODE(info.st_mode) & 0o002:
        raise ProbeError("unsafe_command_queue")
    data = (json.dumps(value, sort_keys=True, separators=(",", ":")) + "\n").encode()
    fd = os.open(path, os.O_WRONLY | os.O_CREAT | os.O_EXCL | getattr(os, "O_NOFOLLOW", 0), 0o640)
    try:
        os.write(fd, data)
        os.fsync(fd)
    finally:
        os.close(fd)
 def _status(database: Path, command_id: str) -> str | None:
    with sqlite3.connect(f"file:{database}?mode=ro", uri=True, timeout=2) as db:
        row = db.execute("SELECT status FROM commands WHERE command_id=?", (command_id,)).fetchone()
    return None if row is None else str(row[0])
 def _markers(path: Path, command_id: str) -> dict[str, int]:
    result = {name: 0 for name in MARKERS}
    with path.open("r", encoding="utf-8") as handle:
        for line in handle:
            try:
                item = json.loads(line)
            except json.JSONDecodeError:
                continue
            if item.get("command_id") == command_id and item.get("marker") in result:
                result[item["marker"]] += 1
    return result
 def run(config_path: Path) -> int:
    config = load_config(config_path)
    endpoint = config["gateway_event_url"]
    gateway_base(endpoint)
    token = read_token(Path(config["gateway_token_file"]))
    command_id = "health-codex-" + uuid.uuid4().hex
    version = state_version(endpoint, config["project_key"], token)
    command = {
        "schema_version": 1,
        "command_id": command_id,
        "project_key": config["project_key"],
        "work_unit_id": command_id,
        "contract_commit": config["contract_commit"],
        "state_version": version,
        "execution": {"kind": "acceptance_probe", "input": "fixed-side-effect-free-health-probe-v1"},
    }
    _write_command(Path(config["codex_command_queue"]) / f"{command_id}.json", command)
    deadline = time.monotonic() + int(config["codex_timeout_seconds"])
    database = Path(config["codex_database_path"])
    audit = Path(config["codex_audit_log"])
    while time.monotonic() < deadline:
        if database.exists() and audit.exists() and _status(database, command_id) == "COMPLETED":
            counts = _markers(audit, command_id)
            if all(counts[name] == 1 for name in MARKERS):
                for suffix in ("started", "ended"):
                    _, event = request_json(endpoint + "/" + command_id + ":" + suffix, token)
                    if event.get("event_id") != command_id + ":" + suffix:
                        raise ProbeError("gateway_lifecycle_missing")
                atomic_receipt(
                    Path(config["codex_receipt_path"]),
                    {
                        "subject": "codex",
                        "health": "HEALTHY",
                        "live_probe_succeeded": True,
                        "observed_at_epoch": int(time.time()),
                        "command_id": command_id,
                        "lifecycle_complete": True,
                        "duplicate_side_effects": False,
                    },
                )
                return 0
        time.sleep(float(config["poll_interval_seconds"]))
    raise ProbeError("codex_lifecycle_not_proven")
 def main() -> int:
    parser = argparse.ArgumentParser()
    parser.add_argument("--config", required=True, type=Path)
    args = parser.parse_args()
    try:
        return run(args.config)
    except (KeyError, OSError, ProbeError, sqlite3.Error, ValueError):
        return 1
 if __name__ == "__main__":
    raise SystemExit(main())
@@ -1,940 +0,0 @@
 #!/usr/bin/env python3
 """Durable GPT + one-executor quorum and failover controller.
 The controller persists decisions before dispatch and never stores credentials.
 Health is accepted only from metadata-only, live-probe receipts.
 """
 from __future__ import annotations
 import argparse
 import hashlib
 import json
 import os
 import sqlite3
 import stat
 import subprocess
 import tempfile
 import time
 import urllib.error
 import urllib.request
 import urllib.parse
 from dataclasses import dataclass
 from enum import Enum
 from pathlib import Path
 from typing import Any, Callable, Iterable, Mapping
 ALLOWED_LIFECYCLE_EVENTS = frozenset(
    {
        "TASK_DISPATCHED",
        "TASK_STARTED",
        "TASK_ENDED",
        "DECISION_APPLIED",
        "PROJECT_SUMMARY",
        "PROJECT_COMPLETED",
    }
 )
 EXECUTORS = ("openclaw", "codex")
 SECRET_KEYS = frozenset(
    {
        "secret",
        "token",
        "password",
        "api_key",
        "authorization",
        "credential",
        "private_key",
        "environment",
        "env",
        "argv",
        "stdout",
        "stderr",
    }
 )
 class Health(str, Enum):
    HEALTHY = "HEALTHY"
    UNHEALTHY = "UNHEALTHY"
    UNKNOWN = "UNKNOWN"
 class ControllerError(RuntimeError):
    """Fail-closed controller error."""
 class EventConflict(ControllerError):
    """An event id was reused with a different payload."""
 class StaleStateVersion(ControllerError):
    """A dispatch attempted to overwrite a newer task state."""
@dataclass(frozen=True)
 class QuorumDecision:
    allowed: bool
    priority: str
    status: str
    active_executor: str | None
    standby_executor: str | None
    repair_required: bool
 def canonical_json(value: Any) -> str:
    return json.dumps(value, sort_keys=True, separators=(",", ":"), ensure_ascii=True)
 def digest_json(value: Any) -> str:
    return hashlib.sha256(canonical_json(value).encode("utf-8")).hexdigest()
 def _contains_forbidden_key(value: Any) -> bool:
    if isinstance(value, Mapping):
        for key, child in value.items():
            normalized = str(key).strip().lower().replace("-", "_")
            if normalized in SECRET_KEYS or _contains_forbidden_key(child):
                return True
    elif isinstance(value, list):
        return any(_contains_forbidden_key(item) for item in value)
    return False
 def validate_metadata_only(value: Mapping[str, Any]) -> None:
    if _contains_forbidden_key(value):
        raise ControllerError("secret-bearing or output-bearing fields are forbidden")
 def atomic_json_write(path: Path, value: Mapping[str, Any], mode: int = 0o600) -> None:
    path.parent.mkdir(parents=True, exist_ok=True, mode=0o700)
    validate_secure_directory(path.parent)
    fd, temporary = tempfile.mkstemp(prefix=f".{path.name}.", dir=str(path.parent))
    try:
        os.fchmod(fd, mode)
        with os.fdopen(fd, "w", encoding="utf-8") as handle:
            handle.write(canonical_json(value))
            handle.write("\n")
            handle.flush()
            os.fsync(handle.fileno())
        os.replace(temporary, path)
    finally:
        if os.path.exists(temporary):
            os.unlink(temporary)
 def validate_secure_directory(path: Path, *, expected_uid: int | None = None) -> None:
    info = path.lstat()
    if stat.S_ISLNK(info.st_mode) or not stat.S_ISDIR(info.st_mode):
        raise ControllerError(f"unsafe directory type rejected: {path}")
    if os.name == "posix" and path.resolve(strict=True) != Path(os.path.abspath(path)):
        raise ControllerError(f"symlinked directory component rejected: {path}")
    if os.name == "posix" and info.st_mode & 0o022:
        raise ControllerError(f"group/world-writable directory rejected: {path}")
    if expected_uid is not None and info.st_uid != expected_uid:
        raise ControllerError(f"unexpected owner for directory {path}")
 def read_secure_json(path: Path, *, expected_uid: int | None = None) -> dict[str, Any]:
    info = path.lstat()
    if stat.S_ISLNK(info.st_mode):
        raise ControllerError(f"symlink rejected: {path}")
    if not stat.S_ISREG(info.st_mode):
        raise ControllerError(f"not a regular file: {path}")
    if os.name == "posix" and info.st_mode & 0o022:
        raise ControllerError(f"group/world-writable file rejected: {path}")
    validate_secure_directory(path.parent, expected_uid=expected_uid)
    if os.name == "posix" and path.resolve(strict=True) != Path(os.path.abspath(path)):
        raise ControllerError(f"symlinked file component rejected: {path}")
    if expected_uid is not None and info.st_uid != expected_uid:
        raise ControllerError(f"unexpected owner for {path}")
    with path.open("r", encoding="utf-8") as handle:
        value = json.load(handle)
    if not isinstance(value, dict):
        raise ControllerError(f"JSON object required: {path}")
    validate_metadata_only(value)
    return value
 def read_gateway_token(path: Path) -> str:
    if not path.is_absolute():
        raise ControllerError("gateway token path must be absolute")
    try:
        info = path.lstat()
    except FileNotFoundError as exc:
        raise ControllerError("gateway token unavailable") from exc
    if stat.S_ISLNK(info.st_mode) or not stat.S_ISREG(info.st_mode):
        raise ControllerError("gateway token must be a regular non-symlink file")
    if stat.S_IMODE(info.st_mode) & ~0o640:
        raise ControllerError("gateway token mode exceeds 0640")
    if os.name == "posix" and info.st_uid not in {0, os.geteuid()}:
        raise ControllerError("gateway token owner is not approved")
    if os.name == "posix" and path.resolve(strict=True) != Path(os.path.abspath(path)):
        raise ControllerError("gateway token has a symlinked path component")
    flags = os.O_RDONLY | getattr(os, "O_NOFOLLOW", 0)
    fd = os.open(path, flags)
    try:
        opened = os.fstat(fd)
        if not stat.S_ISREG(opened.st_mode) or opened.st_ino != info.st_ino or opened.st_dev != info.st_dev:
            raise ControllerError("gateway token changed during validation")
        data = os.read(fd, 16_385)
        if os.read(fd, 1):
            raise ControllerError("gateway token too large")
    finally:
        os.close(fd)
    if not data or len(data) > 16_384 or b"\n" in data or b"\r" in data:
        raise ControllerError("gateway token format rejected")
    try:
        token = data.decode("utf-8")
    except UnicodeDecodeError as exc:
        raise ControllerError("gateway token encoding rejected") from exc
    if not token:
        raise ControllerError("gateway token unavailable")
    return token
 def validate_health_receipt(
    receipt: Mapping[str, Any],
    subject: str,
    *,
    now: float | None = None,
    max_age_seconds: int = 60,
 ) -> Health:
    validate_metadata_only(receipt)
    if receipt.get("subject") != subject:
        return Health.UNKNOWN
    try:
        health = Health(str(receipt.get("health", "UNKNOWN")))
    except ValueError:
        return Health.UNKNOWN
    if health is not Health.HEALTHY:
        return health
    if receipt.get("live_probe_succeeded") is not True:
        return Health.UNKNOWN
    observed_at = receipt.get("observed_at_epoch")
    if not isinstance(observed_at, (int, float)):
        return Health.UNKNOWN
    now = time.time() if now is None else now
    if observed_at > now + 5 or now - observed_at > max_age_seconds:
        return Health.UNKNOWN
    if subject == "central_gpt":
        workspace = receipt.get("workspace_agent")
        responses = receipt.get("responses_api")
        workspace_ok = isinstance(workspace, Mapping) and workspace.get("validated_live") is True
        responses_ok = isinstance(responses, Mapping) and responses.get("validated_live") is True
        if not (workspace_ok or responses_ok):
            return Health.UNKNOWN
    return Health.HEALTHY
 def choose_executor(workload: str) -> str:
    if workload in {"complex_code", "security_boundary", "architecture", "complex_code_security_architecture"}:
        return "codex"
    return "openclaw"
 def evaluate_quorum(
    central_gpt: Health,
    openclaw: Health,
    codex: Health,
    workload: str,
 ) -> QuorumDecision:
    if central_gpt is not Health.HEALTHY:
        status = "BLOCKED_CENTRAL_GPT_UNKNOWN" if central_gpt is Health.UNKNOWN else "BLOCKED_CENTRAL_GPT_UNHEALTHY"
        return QuorumDecision(False, "P0", status, None, None, True)
    healthy = [name for name, health in (("openclaw", openclaw), ("codex", codex)) if health is Health.HEALTHY]
    if not healthy:
        has_unknown = Health.UNKNOWN in {openclaw, codex}
        status = "BLOCKED_EXECUTOR_HEALTH_UNKNOWN" if has_unknown else "BLOCKED_NO_HEALTHY_EXECUTOR"
        return QuorumDecision(False, "P1", status, None, None, True)
    if len(healthy) == 1:
        active = healthy[0]
        standby = "codex" if active == "openclaw" else "openclaw"
        return QuorumDecision(True, "P3", "EXECUTOR_QUORUM_PASS", active, standby, False)
    active = choose_executor(workload)
    standby = "codex" if active == "openclaw" else "openclaw"
    return QuorumDecision(True, "NONE", "EXECUTOR_QUORUM_PASS", active, standby, False)
 class QuorumStore:
    def __init__(self, path: Path):
        self.path = path
        path.parent.mkdir(parents=True, exist_ok=True, mode=0o700)
        expected_uid = os.geteuid() if os.name == "posix" else None
        validate_secure_directory(path.parent, expected_uid=expected_uid)
        if path.exists() or path.is_symlink():
            info = path.lstat()
            if stat.S_ISLNK(info.st_mode) or not stat.S_ISREG(info.st_mode) or info.st_nlink != 1:
                raise ControllerError(f"unsafe database path rejected: {path}")
            if expected_uid is not None and info.st_uid != expected_uid:
                raise ControllerError(f"unexpected database owner: {path}")
        self.db = sqlite3.connect(str(path), isolation_level=None, timeout=30)
        if os.name == "posix":
            path.chmod(0o600)
        self.db.row_factory = sqlite3.Row
        self.db.execute("PRAGMA journal_mode=WAL")
        self.db.execute("PRAGMA synchronous=FULL")
        self._initialize()
    def close(self) -> None:
        self.db.close()
    def _initialize(self) -> None:
        self.db.executescript(
            """
            CREATE TABLE IF NOT EXISTS events (
                event_id TEXT PRIMARY KEY,
                payload_digest TEXT NOT NULL,
                payload_json TEXT NOT NULL,
                source TEXT NOT NULL,
                status TEXT NOT NULL,
                created_at REAL NOT NULL
            );
            CREATE TABLE IF NOT EXISTS tasks (
                work_unit TEXT PRIMARY KEY,
                contract_commit TEXT NOT NULL,
                event_id TEXT NOT NULL,
                state_version INTEGER NOT NULL,
                workload TEXT NOT NULL,
                status TEXT NOT NULL,
                active_executor TEXT,
                standby_executor TEXT,
                checkpoint TEXT,
                lease_owner TEXT,
                lease_until REAL,
                interrupted_executor TEXT,
                recovery_priority TEXT NOT NULL,
                updated_at REAL NOT NULL
            );
            CREATE TABLE IF NOT EXISTS lifecycle (
                sequence INTEGER PRIMARY KEY AUTOINCREMENT,
                work_unit TEXT NOT NULL,
                event_type TEXT NOT NULL,
                executor TEXT,
                state_version INTEGER NOT NULL,
                metadata_json TEXT NOT NULL,
                created_at REAL NOT NULL
            );
            CREATE TABLE IF NOT EXISTS effects (
                effect_id TEXT PRIMARY KEY,
                work_unit TEXT NOT NULL,
                payload_digest TEXT NOT NULL,
                status TEXT NOT NULL,
                created_at REAL NOT NULL
            );
            CREATE TABLE IF NOT EXISTS outbox (
                outbox_id TEXT PRIMARY KEY,
                destination TEXT NOT NULL,
                payload_digest TEXT NOT NULL,
                payload_json TEXT NOT NULL,
                status TEXT NOT NULL,
                attempts INTEGER NOT NULL DEFAULT 0,
                next_attempt REAL NOT NULL,
                created_at REAL NOT NULL
            );
            CREATE TABLE IF NOT EXISTS blockers (
                blocker_key TEXT PRIMARY KEY,
                occurrences INTEGER NOT NULL,
                status TEXT NOT NULL,
                updated_at REAL NOT NULL
            );
            CREATE TABLE IF NOT EXISTS observed_refs (
                ref_name TEXT PRIMARY KEY,
                oid TEXT NOT NULL,
                updated_at REAL NOT NULL
            );
            CREATE TABLE IF NOT EXISTS summaries (
                work_unit TEXT PRIMARY KEY,
                summary_digest TEXT NOT NULL,
                summary_json TEXT NOT NULL,
                created_at REAL NOT NULL
            );
            CREATE UNIQUE INDEX IF NOT EXISTS lifecycle_once
            ON lifecycle(work_unit, event_type, COALESCE(executor, ''), state_version);
            """
        )
    def ingest_event(self, event_id: str, payload: Mapping[str, Any], source: str) -> bool:
        validate_metadata_only(payload)
        encoded = canonical_json(payload)
        digest = hashlib.sha256(encoded.encode("utf-8")).hexdigest()
        existing = self.db.execute("SELECT payload_digest FROM events WHERE event_id=?", (event_id,)).fetchone()
        if existing:
            if existing["payload_digest"] != digest:
                self.db.execute("UPDATE events SET status='DEAD_LETTER_CONFLICT' WHERE event_id=?", (event_id,))
                raise EventConflict(f"event id payload conflict: {event_id}")
            return False
        self.db.execute(
            "INSERT INTO events VALUES(?,?,?,?,?,?)",
            (event_id, digest, encoded, source, "ACCEPTED", time.time()),
        )
        return True
    def upsert_task(
        self,
        *,
        work_unit: str,
        contract_commit: str,
        event_id: str,
        state_version: int,
        workload: str,
        decision: QuorumDecision,
        checkpoint: str | None = None,
    ) -> str:
        now = time.time()
        self.db.execute("BEGIN IMMEDIATE")
        try:
            current = self.db.execute("SELECT * FROM tasks WHERE work_unit=?", (work_unit,)).fetchone()
            if current and state_version < current["state_version"]:
                raise StaleStateVersion(f"stale state version for {work_unit}")
            interrupted: str | None = None
            recovery_priority = decision.priority
            status = "BLOCKED"
            if decision.allowed:
                status = "READY"
                if current and current["active_executor"] and current["active_executor"] != decision.active_executor:
                    interrupted = current["active_executor"]
                    recovery_priority = "P2"
                    status = "INTERRUPTED_AGENT_FAILOVER"
                    if checkpoint is None:
                        checkpoint = current["checkpoint"]
            if current:
                if contract_commit != current["contract_commit"] or event_id != current["event_id"]:
                    raise ControllerError("failover must preserve contract_commit and event_id")
                if (
                    current["state_version"] == state_version
                    and current["active_executor"] == decision.active_executor
                    and current["status"] in {"RUNNING", "COMPLETED"}
                ):
                    self.db.execute("COMMIT")
                    return current["status"]
                self.db.execute(
                    """UPDATE tasks SET state_version=?, workload=?, status=?, active_executor=?,
                    standby_executor=?, checkpoint=?, interrupted_executor=?, recovery_priority=?, updated_at=? WHERE work_unit=?""",
                    (
                        state_version,
                        workload,
                        status,
                        decision.active_executor,
                        decision.standby_executor,
                        checkpoint,
                        interrupted,
                        recovery_priority,
                        now,
                        work_unit,
                    ),
                )
            else:
                self.db.execute(
                    "INSERT INTO tasks VALUES(?,?,?,?,?,?,?,?,?,?,?,?,?,?)",
                    (
                        work_unit,
                        contract_commit,
                        event_id,
                        state_version,
                        workload,
                        status,
                        decision.active_executor,
                        decision.standby_executor,
                        checkpoint,
                        None,
                        None,
                        interrupted,
                        recovery_priority,
                        now,
                    ),
                )
            self.db.execute("COMMIT")
            return status
        except Exception:
            self.db.execute("ROLLBACK")
            raise
    def claim_lease(self, work_unit: str, worker: str, ttl_seconds: int, now: float | None = None) -> bool:
        now = time.time() if now is None else now
        self.db.execute("BEGIN IMMEDIATE")
        try:
            task = self.db.execute(
                "SELECT active_executor, status, lease_owner, lease_until FROM tasks WHERE work_unit=?",
                (work_unit,),
            ).fetchone()
            if not task:
                raise ControllerError(f"unknown work unit: {work_unit}")
            if worker != task["active_executor"]:
                self.db.execute("ROLLBACK")
                return False
            if task["status"] in {"BLOCKED", "COMPLETED"}:
                self.db.execute("ROLLBACK")
                return False
            lease_live = task["lease_owner"] and task["lease_until"] and task["lease_until"] > now
            if lease_live and task["lease_owner"] != worker:
                self.db.execute("ROLLBACK")
                return False
            self.db.execute(
                "UPDATE tasks SET lease_owner=?, lease_until=?, status='RUNNING', updated_at=? WHERE work_unit=?",
                (worker, now + ttl_seconds, now, work_unit),
            )
            self.db.execute("COMMIT")
            return True
        except Exception:
            if self.db.in_transaction:
                self.db.execute("ROLLBACK")
            raise
    def dispatch_task(
        self,
        *,
        work_unit: str,
        contract_commit: str,
        event_id: str,
        state_version: int,
        workload: str,
        decision: QuorumDecision,
        checkpoint: str | None = None,
    ) -> str:
        status = self.upsert_task(
            work_unit=work_unit,
            contract_commit=contract_commit,
            event_id=event_id,
            state_version=state_version,
            workload=workload,
            decision=decision,
            checkpoint=checkpoint,
        )
        if decision.allowed and status != "COMPLETED":
            metadata = {
                "contract_commit": contract_commit,
                "event_id": event_id,
                "status": status,
            }
            self.append_lifecycle(work_unit, "TASK_DISPATCHED", decision.active_executor, state_version, metadata)
            self.enqueue(
                f"dispatch:{work_unit}:{state_version}:{decision.active_executor}",
                "gateway",
                {"event_type": "TASK_DISPATCHED", "work_unit": work_unit, **metadata},
            )
        return status
    def start_task(self, work_unit: str, executor: str, state_version: int, lease_ttl_seconds: int) -> bool:
        if not self.claim_lease(work_unit, executor, lease_ttl_seconds):
            return False
        self.append_lifecycle(work_unit, "TASK_STARTED", executor, state_version, {"status": "RUNNING"})
        self.enqueue(
            f"started:{work_unit}:{state_version}:{executor}",
            "gateway",
            {"event_type": "TASK_STARTED", "work_unit": work_unit, "executor": executor, "state_version": state_version},
        )
        return True
    def end_task(
        self,
        work_unit: str,
        executor: str,
        state_version: int,
        result: str,
        summary: Mapping[str, Any],
        *,
        project_completed: bool = False,
    ) -> None:
        validate_metadata_only(summary)
        self.db.execute("BEGIN IMMEDIATE")
        try:
            task = self.db.execute("SELECT active_executor,state_version FROM tasks WHERE work_unit=?", (work_unit,)).fetchone()
            if not task or task["active_executor"] != executor:
                raise ControllerError("only the active executor may end a task")
            if task["state_version"] != state_version:
                raise StaleStateVersion(f"stale state version for {work_unit}")
            self.db.execute(
                "UPDATE tasks SET status='COMPLETED', lease_owner=NULL, lease_until=NULL, updated_at=? WHERE work_unit=?",
                (time.time(), work_unit),
            )
            self.append_lifecycle(work_unit, "TASK_ENDED", executor, state_version, {"result": result})
            self.append_lifecycle(work_unit, "DECISION_APPLIED", executor, state_version, {"result": result})
            self.append_lifecycle(work_unit, "PROJECT_SUMMARY", executor, state_version, summary)
            if project_completed:
                self.append_lifecycle(work_unit, "PROJECT_COMPLETED", executor, state_version, {"result": result})
            self.persist_summary(work_unit, summary)
            outbound = (
                ("TASK_ENDED", {"result": result}),
                ("DECISION_APPLIED", {"result": result}),
                ("PROJECT_SUMMARY", dict(summary)),
            )
            for event_type, metadata in outbound:
                self.enqueue(
                    f"{event_type.lower()}:{work_unit}:{state_version}:{executor}",
                    "gateway",
                    {
                        "event_type": event_type,
                        "work_unit": work_unit,
                        "executor": executor,
                        "state_version": state_version,
                        **metadata,
                    },
                )
            if project_completed:
                self.enqueue(
                    f"project_completed:{work_unit}:{state_version}:{executor}",
                    "gateway",
                    {
                        "event_type": "PROJECT_COMPLETED",
                        "work_unit": work_unit,
                        "executor": executor,
                        "state_version": state_version,
                        "result": result,
                    },
                )
            self.db.execute("COMMIT")
        except Exception:
            if self.db.in_transaction:
                self.db.execute("ROLLBACK")
            raise
    def record_effect(self, effect_id: str, work_unit: str, payload: Mapping[str, Any]) -> bool:
        validate_metadata_only(payload)
        digest = digest_json(payload)
        existing = self.db.execute("SELECT payload_digest FROM effects WHERE effect_id=?", (effect_id,)).fetchone()
        if existing:
            if existing["payload_digest"] != digest:
                raise EventConflict(f"effect id payload conflict: {effect_id}")
            return False
        self.db.execute(
            "INSERT INTO effects VALUES(?,?,?,?,?)",
            (effect_id, work_unit, digest, "COMPLETED", time.time()),
        )
        return True
    def append_lifecycle(
        self,
        work_unit: str,
        event_type: str,
        executor: str | None,
        state_version: int,
        metadata: Mapping[str, Any],
    ) -> None:
        if event_type not in ALLOWED_LIFECYCLE_EVENTS:
            raise ControllerError(f"unsupported lifecycle event: {event_type}")
        validate_metadata_only(metadata)
        self.db.execute(
            "INSERT OR IGNORE INTO lifecycle(work_unit,event_type,executor,state_version,metadata_json,created_at) VALUES(?,?,?,?,?,?)",
            (work_unit, event_type, executor, state_version, canonical_json(metadata), time.time()),
        )
    def persist_summary(self, work_unit: str, summary: Mapping[str, Any]) -> None:
        validate_metadata_only(summary)
        encoded = canonical_json(summary)
        digest = hashlib.sha256(encoded.encode("utf-8")).hexdigest()
        self.db.execute(
            "INSERT OR REPLACE INTO summaries VALUES(?,?,?,?)",
            (work_unit, digest, encoded, time.time()),
        )
    def enqueue(self, outbox_id: str, destination: str, payload: Mapping[str, Any]) -> bool:
        validate_metadata_only(payload)
        encoded = canonical_json(payload)
        digest = hashlib.sha256(encoded.encode("utf-8")).hexdigest()
        existing = self.db.execute("SELECT payload_digest FROM outbox WHERE outbox_id=?", (outbox_id,)).fetchone()
        if existing:
            if existing["payload_digest"] != digest:
                raise EventConflict(f"outbox id payload conflict: {outbox_id}")
            return False
        self.db.execute(
            "INSERT INTO outbox VALUES(?,?,?,?,?,?,?,?)",
            (outbox_id, destination, digest, encoded, "PENDING", 0, time.time(), time.time()),
        )
        return True
    def record_blocker(self, blocker_key: str, threshold: int) -> str:
        now = time.time()
        row = self.db.execute("SELECT occurrences FROM blockers WHERE blocker_key=?", (blocker_key,)).fetchone()
        occurrences = (row["occurrences"] if row else 0) + 1
        status = "DEAD_LETTER" if occurrences > threshold else "RETRY"
        self.db.execute(
            "INSERT INTO blockers VALUES(?,?,?,?) ON CONFLICT(blocker_key) DO UPDATE SET occurrences=excluded.occurrences,status=excluded.status,updated_at=excluded.updated_at",
            (blocker_key, occurrences, status, now),
        )
        return status
    def deliver_outbox(self, sender: Callable[[Mapping[str, Any]], None], max_attempts: int, now: float | None = None) -> tuple[int, int]:
        now = time.time() if now is None else now
        sent = 0
        failed = 0
        rows = self.db.execute(
            "SELECT * FROM outbox WHERE status='PENDING' AND next_attempt<=? ORDER BY created_at,outbox_id LIMIT 100",
            (now,),
        ).fetchall()
        for row in rows:
            payload = json.loads(row["payload_json"])
            validate_metadata_only(payload)
            try:
                sender(payload)
            except Exception:
                attempts = row["attempts"] + 1
                status = "DEAD_LETTER" if attempts >= max_attempts else "PENDING"
                delay = min(300, 2 ** min(attempts, 8))
                self.db.execute(
                    "UPDATE outbox SET attempts=?,status=?,next_attempt=? WHERE outbox_id=?",
                    (attempts, status, now + delay, row["outbox_id"]),
                )
                failed += 1
            else:
                self.db.execute(
                    "UPDATE outbox SET attempts=attempts+1,status='SENT',next_attempt=? WHERE outbox_id=?",
                    (now, row["outbox_id"]),
                )
                sent += 1
        return sent, failed
    def observe_ref(self, ref_name: str, oid: str) -> bool:
        row = self.db.execute("SELECT oid FROM observed_refs WHERE ref_name=?", (ref_name,)).fetchone()
        changed = row is not None and row["oid"] != oid
        self.db.execute(
            "INSERT INTO observed_refs VALUES(?,?,?) ON CONFLICT(ref_name) DO UPDATE SET oid=excluded.oid,updated_at=excluded.updated_at",
            (ref_name, oid, time.time()),
        )
        return changed
 def spool_after_http_failure(
    sender: Callable[[Mapping[str, Any]], None],
    payload: Mapping[str, Any],
    spool_dir: Path,
    event_id: str,
 ) -> bool:
    validate_metadata_only(payload)
    try:
        sender(payload)
        return True
    except Exception:
        atomic_json_write(spool_dir / f"{event_id}.json", {"event_id": event_id, "payload": dict(payload)})
        return False
 def poll_file_inbox(store: QuorumStore, inbox_dir: Path) -> int:
    if not inbox_dir.exists():
        return 0
    validate_secure_directory(inbox_dir, expected_uid=os.geteuid() if os.name == "posix" else None)
    count = 0
    for path in sorted(inbox_dir.glob("*.json")):
        if path.is_symlink():
            raise ControllerError(f"symlink inbox entry rejected: {path}")
        item = read_secure_json(path, expected_uid=os.geteuid() if os.name == "posix" else None)
        event_id = item.get("event_id")
        payload = item.get("payload")
        if not isinstance(event_id, str) or not isinstance(payload, dict):
            raise ControllerError(f"invalid inbox envelope: {path}")
        if store.ingest_event(event_id, payload, "file-inbox"):
            count += 1
        path.unlink()
    return count
 def resolve_git_ref(repo: Path, ref_name: str) -> str:
    if not repo.is_absolute():
        raise ControllerError("Git repository path must be absolute")
    info = repo.lstat()
    if stat.S_ISLNK(info.st_mode) or not stat.S_ISDIR(info.st_mode):
        raise ControllerError("Git repository must be a non-symlink directory")
    if os.name == "posix" and repo.resolve(strict=True) != Path(os.path.abspath(repo)):
        raise ControllerError("Git repository has a symlinked path component")
    if os.name == "posix" and info.st_mode & 0o002:
        raise ControllerError("world-writable Git repository rejected")
    try:
        completed = subprocess.run(
            [
                "git",
                "-c",
                f"safe.directory={repo}",
                "-C",
                str(repo),
                "rev-parse",
                "--verify",
                ref_name,
            ],
            check=True,
            stdout=subprocess.PIPE,
            stderr=subprocess.DEVNULL,
            text=True,
            timeout=10,
        )
    except (OSError, subprocess.CalledProcessError, subprocess.TimeoutExpired) as exc:
        raise ControllerError("Git ref unavailable") from exc
    oid = completed.stdout.strip()
    if len(oid) != 40 or any(character not in "0123456789abcdef" for character in oid.lower()):
        raise ControllerError("invalid Git object id")
    return oid
 def poll_git_refs(
    store: QuorumStore,
    repo: Path,
    refs: Iterable[str],
    resolver: Callable[[Path, str], str] = resolve_git_ref,
 ) -> int:
    changed = 0
    for ref_name in refs:
        if not ref_name.startswith("refs/") or ".." in ref_name or ref_name.endswith("/"):
            raise ControllerError(f"invalid Git ref: {ref_name}")
        oid = resolver(repo, ref_name)
        if store.observe_ref(ref_name, oid):
            event_id = f"gitref:{ref_name}:{oid}"
            if store.ingest_event(event_id, {"ref": ref_name, "oid": oid}, "git-ref-poll"):
                changed += 1
    return changed
 def responses_fallback_allowed(
    workspace_agent_ok: bool,
    receipt: Mapping[str, Any],
    *,
    now: float | None = None,
 ) -> bool:
    validate_metadata_only(receipt)
    if workspace_agent_ok:
        return False
    responses = receipt.get("responses_api")
    return (
        validate_health_receipt(receipt, "central_gpt", now=now) is Health.HEALTHY
        and isinstance(responses, Mapping)
        and responses.get("validated_live") is True
    )
 def _validate_gateway_url(url: str) -> None:
    parsed = urllib.parse.urlparse(url)
    if (
        parsed.scheme != "http"
        or parsed.hostname != "127.0.0.1"
        or parsed.port != 18920
        or parsed.path != "/v1/orchestrator/events"
        or parsed.params
        or parsed.query
        or parsed.fragment
    ):
        raise ControllerError("gateway endpoint must be the fixed loopback endpoint")
 def get_gateway_state(url: str, project_key: str, token: str, timeout: float = 5.0) -> int:
    _validate_gateway_url(url)
    state_url = url.removesuffix("/events") + "/state/" + urllib.parse.quote(project_key, safe="")
    request = urllib.request.Request(
        state_url,
        headers={"Authorization": f"Bearer {token}"},
        method="GET",
    )
    try:
        with urllib.request.urlopen(request, timeout=timeout) as response:
            payload = json.loads(response.read(262_145))
    except urllib.error.HTTPError as exc:
        if exc.code == 404:
            return 0
        raise ControllerError(f"gateway state rejected HTTP {exc.code}") from exc
    except (urllib.error.URLError, TimeoutError, json.JSONDecodeError) as exc:
        raise ControllerError("gateway state unavailable") from exc
    version = payload.get("state_version") if isinstance(payload, Mapping) else None
    if not isinstance(version, int) or version < 0:
        raise ControllerError("gateway state version invalid")
    return version
 def post_json(url: str, payload: Mapping[str, Any], token: str, timeout: float = 5.0) -> None:
    validate_metadata_only(payload)
    _validate_gateway_url(url)
    request = urllib.request.Request(
        url,
        data=canonical_json(payload).encode("utf-8"),
        headers={"Content-Type": "application/json", "Authorization": f"Bearer {token}"},
        method="POST",
    )
    try:
        with urllib.request.urlopen(request, timeout=timeout) as response:
            if response.status < 200 or response.status >= 300:
                raise ControllerError(f"gateway rejected HTTP {response.status}")
    except urllib.error.HTTPError as exc:
        raise ControllerError(f"gateway rejected HTTP {exc.code}") from exc
    except (urllib.error.URLError, TimeoutError) as exc:
        raise ControllerError("gateway unavailable") from exc
 def run_once(config_path: Path) -> int:
    config = read_secure_json(config_path, expected_uid=0 if os.name == "posix" else None)
    gateway_url = config["gateway_event_url"]
    token = read_gateway_token(Path(config["gateway_token_file"]))
    state_version = get_gateway_state(gateway_url, config["project_key"], token)
    store = QuorumStore(Path(config["database_path"]))
    try:
        receipts: dict[str, Health] = {}
        for subject in ("central_gpt", "openclaw", "codex"):
            receipt_path = Path(config["health_receipts"][subject])
            try:
                receipt = read_secure_json(receipt_path, expected_uid=0 if os.name == "posix" else None)
                receipts[subject] = validate_health_receipt(
                    receipt,
                    subject,
                    max_age_seconds=int(config.get("max_health_age_seconds", 60)),
                )
            except (FileNotFoundError, ControllerError, json.JSONDecodeError):
                receipts[subject] = Health.UNKNOWN
        decision = evaluate_quorum(
            receipts["central_gpt"], receipts["openclaw"], receipts["codex"], config.get("default_workload", "large_repetitive_infrastructure_monitoring")
        )
        quorum_event_id = f"quorum-{time.time_ns()}"
        store.enqueue(
            quorum_event_id,
            "gateway",
            {
                "schema_version": 1,
                "project_key": config["project_key"],
                "contract_commit": config["contract_commit"],
                "event_id": quorum_event_id,
                "source_agent": "system",
                "event_type": "executor_quorum_evaluated",
                "work_unit_id": "A0-AUTOMATION-QUORUM",
                "status": "READY" if decision.allowed else "BLOCKED",
                "state_version": state_version,
                "implementation_commit": None,
                "result_commit": None,
                "blocker": None if decision.allowed else decision.status,
                "evidence_refs": [],
                "payload": {
                    "allowed": decision.allowed,
                    "priority": decision.priority,
                    "quorum_status": decision.status,
                    "active_executor": decision.active_executor,
                    "standby_executor": decision.standby_executor,
                },
            },
        )
        poll_file_inbox(store, Path(config["file_inbox_path"]))
        git_polling = config.get("git_ref_polling")
        if isinstance(git_polling, Mapping):
            poll_git_refs(store, Path(git_polling["repository"]), git_polling["refs"])
        store.deliver_outbox(
            lambda payload: post_json(gateway_url, payload, token),
            max_attempts=int(config["blocker_retry_threshold"]),
        )
        return 0 if decision.allowed else 2
    finally:
        store.close()
 def build_parser() -> argparse.ArgumentParser:
    parser = argparse.ArgumentParser(description=__doc__)
    parser.add_argument("--config", type=Path, required=True)
    parser.add_argument("--once", action="store_true", required=True)
    return parser
 def main(argv: Iterable[str] | None = None) -> int:
    args = build_parser().parse_args(argv)
    try:
        return run_once(args.config)
    except (ControllerError, KeyError, OSError, sqlite3.Error, json.JSONDecodeError) as error:
        print(f"automation-quorum-controller: fail-closed: {type(error).__name__}", file=os.sys.stderr)
        return 3
 if __name__ == "__main__":
    raise SystemExit(main())
@@ -1,146 +0,0 @@
 #!/usr/bin/env python3
 """Strict metadata-only helpers shared by root health probes."""
 from __future__ import annotations
 import json
 import os
 from pathlib import Path
 import stat
 import tempfile
 from typing import Any
 import urllib.error
 import urllib.parse
 import urllib.request
 class ProbeError(RuntimeError):
    pass
 def secure_regular(path: Path, *, max_mode: int, owners: set[int]) -> os.stat_result:
    if not path.is_absolute():
        raise ProbeError("path_not_absolute")
    info = path.lstat()
    if stat.S_ISLNK(info.st_mode) or not stat.S_ISREG(info.st_mode):
        raise ProbeError("unsafe_file_type")
    if stat.S_IMODE(info.st_mode) & ~max_mode:
        raise ProbeError("unsafe_file_mode")
    if os.name == "posix" and info.st_uid not in owners:
        raise ProbeError("unsafe_file_owner")
    if os.name == "posix" and path.resolve(strict=True) != Path(os.path.abspath(path)):
        raise ProbeError("symlinked_path_component")
    return info
 def load_config(path: Path) -> dict[str, Any]:
    secure_regular(path, max_mode=0o600, owners={0})
    value = json.loads(path.read_text(encoding="utf-8"))
    if not isinstance(value, dict):
        raise ProbeError("config_not_object")
    return value
 def read_token(path: Path) -> str:
    info = secure_regular(path, max_mode=0o640, owners={0, os.geteuid()})
    if info.st_size <= 0 or info.st_size > 16_384:
        raise ProbeError("invalid_token_size")
    fd = os.open(path, os.O_RDONLY | getattr(os, "O_NOFOLLOW", 0))
    try:
        opened = os.fstat(fd)
        if not stat.S_ISREG(opened.st_mode) or opened.st_ino != info.st_ino or opened.st_dev != info.st_dev:
            raise ProbeError("token_changed_during_validation")
        data = os.read(fd, 16_385)
        if os.read(fd, 1):
            raise ProbeError("invalid_token_size")
    finally:
        os.close(fd)
    if b"\n" in data or b"\r" in data:
        raise ProbeError("invalid_token_format")
    try:
        return data.decode("utf-8")
    except UnicodeDecodeError as exc:
        raise ProbeError("invalid_token_encoding") from exc
 def gateway_base(endpoint: str) -> str:
    parsed = urllib.parse.urlparse(endpoint)
    if (
        parsed.scheme != "http"
        or parsed.hostname != "127.0.0.1"
        or parsed.port != 18920
        or parsed.path != "/v1/orchestrator/events"
        or parsed.query
        or parsed.fragment
    ):
        raise ProbeError("invalid_gateway_endpoint")
    return endpoint.removesuffix("/events")
 def request_json(
    url: str,
    token: str,
    *,
    method: str = "GET",
    payload: dict[str, Any] | None = None,
    timeout: float = 10.0,
 ) -> tuple[int, dict[str, Any]]:
    data = None if payload is None else json.dumps(payload, sort_keys=True, separators=(",", ":")).encode()
    headers = {"Authorization": f"Bearer {token}"}
    if data is not None:
        headers["Content-Type"] = "application/json"
    request = urllib.request.Request(url, data=data, headers=headers, method=method)
    try:
        with urllib.request.urlopen(request, timeout=timeout) as response:
            body = json.loads(response.read(262_145))
            if not isinstance(body, dict):
                raise ProbeError("gateway_response_not_object")
            return response.status, body
    except urllib.error.HTTPError as exc:
        if exc.code in {401, 403, 409, 422}:
            raise ProbeError(f"gateway_fail_closed_{exc.code}") from exc
        raise ProbeError(f"gateway_http_{exc.code}") from exc
    except (urllib.error.URLError, TimeoutError, json.JSONDecodeError) as exc:
        raise ProbeError("gateway_unavailable") from exc
 def state_version(endpoint: str, project_key: str, token: str) -> int:
    url = gateway_base(endpoint) + "/state/" + urllib.parse.quote(project_key, safe="")
    try:
        _, value = request_json(url, token)
    except ProbeError as exc:
        if str(exc) == "gateway_http_404":
            return 0
        raise
    version = value.get("state_version")
    if not isinstance(version, int) or version < 0:
        raise ProbeError("invalid_state_version")
    return version
 def atomic_receipt(path: Path, value: dict[str, Any]) -> None:
    if not path.is_absolute() or path.parent != Path("/run/aios-health"):
        raise ProbeError("invalid_receipt_path")
    parent = path.parent
    info = parent.lstat()
    if stat.S_ISLNK(info.st_mode) or not stat.S_ISDIR(info.st_mode):
        raise ProbeError("unsafe_receipt_directory")
    if os.name == "posix" and (info.st_uid != 0 or stat.S_IMODE(info.st_mode) != 0o750):
        raise ProbeError("unsafe_receipt_directory_metadata")
    fd, temporary = tempfile.mkstemp(prefix=f".{path.name}.", dir=parent)
    try:
        # Receipts contain availability metadata only. Give the dedicated
        # aios-quorum group read access so its unprivileged controller can
        # consume them, while retaining root ownership and no public access.
        if os.name == "posix":
            os.fchown(fd, 0, info.st_gid)
        os.fchmod(fd, 0o640)
        with os.fdopen(fd, "w", encoding="utf-8") as handle:
            json.dump(value, handle, sort_keys=True, separators=(",", ":"))
            handle.write("\n")
            handle.flush()
            os.fsync(handle.fileno())
        os.replace(temporary, path)
    finally:
        if os.path.exists(temporary):
            os.unlink(temporary)
@@ -1,24 +0,0 @@
 {
  "$schema": "../schemas/automation-quorum-controller.schema.json",
  "project_key": "v2r1-personal-weixin-production",
  "contract_commit": "cbe33ecab8e9794ba5f318bea43aeaf447a56fbc",
  "database_path": "/var/lib/aios-automation-quorum/controller.sqlite3",
  "file_inbox_path": "/var/lib/aios-automation-quorum/inbox",
  "default_workload": "large_repetitive_infrastructure_monitoring",
  "health_receipts": {
    "central_gpt": "/run/aios-health/central-gpt.json",
    "openclaw": "/run/aios-health/openclaw.json",
    "codex": "/run/aios-health/codex.json"
  },
  "git_ref_polling": {
    "repository": "/data/aios-gitops/repo.git",
    "refs": [
      "refs/heads/task/governance/v2r1-personal-weixin-production-contract-20260717"
    ]
  },
  "blocker_retry_threshold": 3,
  "max_health_age_seconds": 60,
  "lease_ttl_seconds": 120,
  "gateway_event_url": "http://127.0.0.1:18920/v1/orchestrator/events",
  "gateway_token_file": "/etc/aios-orchestrator-gateway/inbound.token"
 }
@@ -1,17 +0,0 @@
 {
  "active_dir": "/data/aios-gitops/workspace/agents/commands/active",
  "completed_dir": "/data/aios-gitops/workspace/agents/commands/completed",
  "dead_letter_dir": "/data/aios-gitops/workspace/agents/commands/dead-letter",
  "database_path": "/var/lib/aios-codex-watcher/state.sqlite3",
  "audit_log_path": "/var/log/aios-codex-watcher/audit.jsonl",
  "gateway_endpoint": "http://127.0.0.1:18920/v1/orchestrator/events",
  "gateway_token_file": "/etc/aios-orchestrator-gateway/inbound.token",
  "contract_commit": "cbe33ecab8e9794ba5f318bea43aeaf447a56fbc",
  "lease_seconds": 60,
  "retry_seconds": 10,
  "max_attempts": 5,
  "poll_seconds": 5,
  "executor_path": null,
  "executor_args": [],
  "executor_timeout_seconds": 1800
 }
@@ -1,17 +0,0 @@
 {
  "project_key": "v2r1-personal-weixin-production",
  "contract_commit": "cbe33ecab8e9794ba5f318bea43aeaf447a56fbc",
  "gateway_event_url": "http://127.0.0.1:18920/v1/orchestrator/events",
  "gateway_token_file": "/etc/aios-orchestrator-gateway/inbound.token",
  "gateway_database_path": "/var/lib/aios-orchestrator-gateway/state.sqlite3",
  "probe_clean_base_commit": "c3984ca2b4388875d2875b4fdfdf95a2984f6ab3",
  "central_receipt_path": "/run/aios-health/central-gpt.json",
  "codex_receipt_path": "/run/aios-health/codex.json",
  "codex_command_queue": "/data/aios-gitops/workspace/agents/commands/active",
  "codex_database_path": "/var/lib/aios-codex-watcher/state.sqlite3",
  "codex_audit_log": "/var/log/aios-codex-watcher/audit.jsonl",
  "central_timeout_seconds": 90,
  "codex_timeout_seconds": 90,
  "cycle_timeout_seconds": 180,
  "poll_interval_seconds": 1
 }
@@ -1,49 +0,0 @@
 {
  "host": "127.0.0.1",
  "port": 18920,
  "database_path": "/var/lib/aios-orchestrator-gateway/state.sqlite3",
  "event_inbox_dir": "/var/lib/aios-orchestrator-gateway/inbox/events",
  "decision_inbox_dir": "/var/lib/aios-orchestrator-gateway/inbox/decisions",
  "dispatch_outbox_dir": "/var/lib/aios-orchestrator-gateway/outbox",
  "openclaw_sync_dir": "/var/lib/aios-orchestrator-gateway/openclaw-sync",
  "command_queue_dir": "/var/lib/aios-orchestrator-gateway/commands",
  "inbound_token_file": "/etc/aios-orchestrator-gateway/inbound.token",
  "callback_token_file": "/etc/aios-orchestrator-gateway/callback.token",
  "contract_commit": "cbe33ecab8e9794ba5f318bea43aeaf447a56fbc",
  "workspace_agent": {
    "trigger_id": "agtch_REPLACE_ME",
    "access_token_file": "/etc/aios-orchestrator-gateway/workspace-agent.token"
  },
  "responses_fallback": {
    "api_key_file": "/etc/aios-orchestrator-gateway/openai-api.key",
    "model": "REPLACE_WITH_APPROVED_MODEL"
  },
  "polling": {
    "interval_seconds": 10,
    "summary_interval_seconds": 300
  },
  "loop_controls": {
    "callback_timeout_seconds": 900,
    "lease_seconds": 60,
    "retry_base_seconds": 10,
    "max_retries": 5,
    "max_same_blocker_retries": 3,
    "max_hops_per_gate": 20
  },
  "openclaw_sync": {
    "endpoint": null,
    "token_file": null
  },
  "project_key": "v2r1-personal-weixin-production",
  "git_poll": {
    "enabled": true,
    "bare_repo": "/data/aios-gitops/repo.git",
    "ref_prefixes": [
      "refs/heads/main",
      "refs/heads/task/governance/",
      "refs/heads/patch/",
      "refs/heads/result/"
    ],
    "interval_seconds": 30
  }
 }
@@ -0,0 +1,23 @@
 FROM python:3.12-slim
 ENV PYTHONDONTWRITEBYTECODE=1 \
    PYTHONUNBUFFERED=1 \
    PYTHONPATH=/opt/aios/app \
    AIOS_V2_HOST=0.0.0.0 \
    AIOS_V2_PORT=18920 \
    AIOS_V2_STATE_DIR=/var/lib/aios-v2
 RUN groupadd --system --gid 12001 aios \
    && useradd --system --uid 12001 --gid aios --home-dir /nonexistent --shell /usr/sbin/nologin aios \
    && install -d -o aios -g aios -m 0750 /opt/aios/app /var/lib/aios-v2
 WORKDIR /opt/aios/app
 COPY --chown=aios:aios active-runtime/app/ /opt/aios/app/
 USER aios:aios
 EXPOSE 18920
 HEALTHCHECK --interval=15s --timeout=5s --start-period=10s --retries=4 \
  CMD python3 -c "import urllib.request; urllib.request.urlopen('http://127.0.0.1:18920/api/v2/health', timeout=3).read()"
 CMD ["python3", "-m", "aios_v2.app"]
@@ -0,0 +1,44 @@
 services:
  aios-v2:
    build:
      context: ../../..
      dockerfile: deploy/aios-v2/docker/Dockerfile
    image: aios-v2:V2R1C0-candidate
    init: true
    restart: unless-stopped
    ports:
      - "127.0.0.1:${AIOS_V2_PORT:-18920}:18920"
    environment:
      AIOS_V2_HOST: 0.0.0.0
      AIOS_V2_PORT: 18920
      AIOS_V2_STATE_DIR: /var/lib/aios-v2
      AIOS_OPENCLAW_URL: ${AIOS_OPENCLAW_URL:-}
      AIOS_OPENCLAW_CREDENTIAL_REF: ${AIOS_OPENCLAW_CREDENTIAL_REF:-}
      AIOS_OPENCLAW_SIGNING_MATERIAL_FILE: ${AIOS_OPENCLAW_SIGNING_MATERIAL_FILE:-}
      AIOS_RELEASE_CHANNEL: candidate
      AIOS_RELEASE_VERSION: V2R1C0-candidate
    volumes:
      - aios-v2-state:/var/lib/aios-v2
      # Mount runtime signing material explicitly when OpenClaw is enabled.
      # The host file must be mode 0600; never store it in this repository.
      # - /etc/aios-v2/openclaw-signing-material:/run/secrets/openclaw-signing-material:ro
    read_only: true
    tmpfs:
      - /tmp:rw,noexec,nosuid,nodev,size=64m
    cap_drop:
      - ALL
    security_opt:
      - no-new-privileges:true
    healthcheck:
      test:
        - CMD
        - python3
        - -c
        - "import urllib.request; urllib.request.urlopen('http://127.0.0.1:18920/api/v2/health', timeout=3).read()"
      interval: 15s
      timeout: 5s
      retries: 4
      start_period: 10s
 volumes:
  aios-v2-state:
@@ -0,0 +1,21 @@
 # AIOS V2 candidate runtime configuration.
 # Copy to /etc/aios-v2/aios-v2.env and keep ownership root:aios, mode 0640.
 AIOS_V2_HOST=127.0.0.1
 AIOS_V2_PORT=18920
 AIOS_V2_STATE_DIR=/var/lib/aios-v2
 # Optional OpenClaw runtime. Leave blank to run with OpenClaw disabled.
 AIOS_OPENCLAW_URL=
 AIOS_OPENCLAW_CREDENTIAL_REF=
 # File path only. The referenced file must be mode 0600 or stricter and contain
 # at least 32 bytes. Never place signing material in this env file or Git.
 AIOS_OPENCLAW_SIGNING_MATERIAL_FILE=
 # Reverse proxy must authenticate users and inject X-AIOS-* identity headers.
 AIOS_TRUST_PROXY_HEADERS=true
 # Candidate deployment metadata exposed through health tooling.
 AIOS_RELEASE_CHANNEL=candidate
 AIOS_RELEASE_VERSION=V2R1C0-candidate
@@ -0,0 +1,86 @@
 #!/usr/bin/env bash
 set -Eeuo pipefail
 BASE_URL="${AIOS_V2_BASE_URL:-http://127.0.0.1:18920}"
 STATE_DIR="${AIOS_V2_STATE_DIR:-/var/lib/aios-v2}"
 CONTROLLER_WT="${AIOS_GITOPS_CONTROLLER_WORKTREE:-/data/aios-gitops/worktrees/controller}"
 SERVICE="${AIOS_V2_SERVICE:-aios-v2}"
 fail() {
  echo "FAIL: $*" >&2
  exit 1
 }
 pass() {
  echo "PASS: $*"
 }
 command -v python3 >/dev/null || fail "python3 not installed"
 if command -v systemctl >/dev/null && systemctl list-unit-files "$SERVICE.service" >/dev/null 2>&1; then
  systemctl is-active --quiet "$SERVICE" || fail "$SERVICE is not active"
  pass "$SERVICE is active"
 fi
 python3 - "$BASE_URL" <<'PY'
 import json
 import sys
 import urllib.request
 base_url = sys.argv[1].rstrip("/")
 with urllib.request.urlopen(f"{base_url}/api/v2/health", timeout=5) as response:
    payload = json.loads(response.read().decode("utf-8"))
 if payload.get("status") not in {"ok", "healthy", "ready"}:
    raise SystemExit(f"unexpected health payload: {payload}")
 print(f"PASS: HTTP health status={payload.get('status')}")
 PY
 [[ -d "$STATE_DIR" ]] || fail "state directory missing: $STATE_DIR"
 [[ -w "$STATE_DIR" ]] || fail "state directory not writable: $STATE_DIR"
 pass "state directory writable"
 if [[ -e "$STATE_DIR/control-plane.sqlite3" ]]; then
  python3 - "$STATE_DIR/control-plane.sqlite3" <<'PY'
 import sqlite3
 import sys
 path = sys.argv[1]
 connection = sqlite3.connect(f"file:{path}?mode=ro", uri=True, timeout=5)
 try:
    row = connection.execute("PRAGMA quick_check").fetchone()
 finally:
    connection.close()
 if not row or row[0] != "ok":
    raise SystemExit(f"SQLite quick_check failed: {row}")
 print("PASS: SQLite quick_check=ok")
 PY
 else
  echo "WARN: control-plane.sqlite3 not created yet"
 fi
 if [[ -d "$CONTROLLER_WT" ]] && git -C "$CONTROLLER_WT" rev-parse --is-inside-work-tree >/dev/null 2>&1; then
  branch="$(git -C "$CONTROLLER_WT" branch --show-current)"
  dirty="$(git -C "$CONTROLLER_WT" status --porcelain=v1)"
  [[ -z "$branch" ]] || fail "GitOps controller attached to branch: $branch"
  [[ -z "$dirty" ]] || fail "GitOps controller workspace is dirty"
  pass "GitOps controller detached and clean"
 fi
 python3 - "$STATE_DIR/executions" <<'PY'
 import json
 import pathlib
 import sys
 root = pathlib.Path(sys.argv[1])
 if not root.exists():
    print("WARN: execution evidence directory not created yet")
    raise SystemExit(0)
 for path in root.glob("*/*.json"):
    value = json.loads(path.read_text(encoding="utf-8"))
    for key in ("execution_id", "tenant_id", "task_id", "executor", "status"):
        if not value.get(key):
            raise SystemExit(f"invalid execution evidence {path}: missing {key}")
 print("PASS: execution evidence files valid")
 PY
 pass "AIOS V2 candidate healthcheck complete"
@@ -0,0 +1,86 @@
 #!/usr/bin/env bash
 set -Eeuo pipefail
 SERVICE="${AIOS_V2_SERVICE:-aios-v2}"
 RELEASE_ROOT="${AIOS_V2_RELEASE_ROOT:-/opt/aios/releases}"
 CURRENT_LINK="${AIOS_V2_CURRENT_LINK:-/opt/aios/current}"
 STATE_DIR="${AIOS_V2_STATE_DIR:-/var/lib/aios-v2}"
 BACKUP_ROOT="${AIOS_V2_BACKUP_ROOT:-/var/backups/aios-v2}"
 TARGET_RELEASE="${1:-}"
 RESTORE_STATE="${AIOS_V2_RESTORE_STATE:-0}"
 fail() {
  echo "ERROR: $*" >&2
  exit 1
 }
 [[ "${EUID}" -eq 0 ]] || fail "rollback must run as root"
 [[ -n "$TARGET_RELEASE" ]] || fail "usage: $0 <release-name-or-absolute-path>"
 if [[ "$TARGET_RELEASE" = /* ]]; then
  TARGET_PATH="$(readlink -f "$TARGET_RELEASE")"
 else
  [[ "$TARGET_RELEASE" != *"/"* && "$TARGET_RELEASE" != "." && "$TARGET_RELEASE" != ".." ]] \
    || fail "invalid release name"
  TARGET_PATH="$(readlink -f "$RELEASE_ROOT/$TARGET_RELEASE")"
 fi
 [[ -d "$TARGET_PATH/active-runtime/app" ]] || fail "target release is invalid: $TARGET_PATH"
 [[ -f "$TARGET_PATH/deploy/aios-v2/healthcheck/check.sh" ]] \
  || fail "target release has no healthcheck"
 PREVIOUS_PATH=""
 if [[ -L "$CURRENT_LINK" ]]; then
  PREVIOUS_PATH="$(readlink -f "$CURRENT_LINK")"
 fi
 STAMP="$(date -u +%Y%m%dT%H%M%SZ)"
 BACKUP_DIR="$BACKUP_ROOT/rollback-$STAMP"
 install -d -m 0750 "$BACKUP_DIR"
 if [[ -d "$STATE_DIR" ]]; then
  tar --xattrs --acls -C "$(dirname "$STATE_DIR")" \
    -czf "$BACKUP_DIR/state-before-rollback.tar.gz" \
    "$(basename "$STATE_DIR")"
 fi
 printf '%s\n' "$PREVIOUS_PATH" > "$BACKUP_DIR/previous-release.txt"
 printf '%s\n' "$TARGET_PATH" > "$BACKUP_DIR/target-release.txt"
 systemctl stop "$SERVICE"
 ln -sfn "$TARGET_PATH" "${CURRENT_LINK}.next"
 mv -Tf "${CURRENT_LINK}.next" "$CURRENT_LINK"
 systemctl daemon-reload
 systemctl start "$SERVICE"
 if AIOS_V2_SERVICE="$SERVICE" \
   AIOS_V2_STATE_DIR="$STATE_DIR" \
   "$TARGET_PATH/deploy/aios-v2/healthcheck/check.sh"; then
  echo "ROLLBACK_COMPLETE target=$TARGET_PATH backup=$BACKUP_DIR"
  exit 0
 fi
 echo "Target release failed healthcheck; restoring previous release." >&2
 systemctl stop "$SERVICE" || true
 if [[ -n "$PREVIOUS_PATH" && -d "$PREVIOUS_PATH" ]]; then
  ln -sfn "$PREVIOUS_PATH" "${CURRENT_LINK}.previous"
  mv -Tf "${CURRENT_LINK}.previous" "$CURRENT_LINK"
 fi
 if [[ "$RESTORE_STATE" == "1" && -f "$BACKUP_DIR/state-before-rollback.tar.gz" ]]; then
  rm -rf "$STATE_DIR"
  tar --xattrs --acls -C "$(dirname "$STATE_DIR")" \
    -xzf "$BACKUP_DIR/state-before-rollback.tar.gz"
 fi
 systemctl daemon-reload
 systemctl start "$SERVICE" || true
 if [[ -n "$PREVIOUS_PATH" && -x "$PREVIOUS_PATH/deploy/aios-v2/healthcheck/check.sh" ]]; then
  AIOS_V2_SERVICE="$SERVICE" \
  AIOS_V2_STATE_DIR="$STATE_DIR" \
  "$PREVIOUS_PATH/deploy/aios-v2/healthcheck/check.sh" || true
 fi
 fail "rollback target failed; previous release restored; evidence=$BACKUP_DIR"
@@ -0,0 +1,37 @@
 [Unit]
 Description=AIOS V2 Control Plane Candidate
 After=network-online.target
 Wants=network-online.target
 [Service]
 Type=simple
 User=aios
 Group=aios
 WorkingDirectory=/opt/aios/current/active-runtime/app
 EnvironmentFile=-/etc/aios-v2/aios-v2.env
 Environment=PYTHONUNBUFFERED=1
 ExecStart=/usr/bin/python3 -m aios_v2.app
 Restart=on-failure
 RestartSec=3
 TimeoutStartSec=30
 TimeoutStopSec=30
 KillSignal=SIGTERM
 NoNewPrivileges=true
 PrivateTmp=true
 ProtectSystem=strict
 ProtectHome=true
 ProtectKernelTunables=true
 ProtectKernelModules=true
 ProtectControlGroups=true
 RestrictSUIDSGID=true
 RestrictRealtime=true
 LockPersonality=true
 MemoryDenyWriteExecute=true
 ReadWritePaths=/var/lib/aios-v2
 StateDirectory=aios-v2
 StateDirectoryMode=0750
 UMask=0027
 LimitNOFILE=65536
 [Install]
 WantedBy=multi-user.target
@@ -13,7 +13,6 @@ Environment=AIOS_GITOPS_TOKEN_FILE=/data/aios-gitops/runtime/secrets/gitops.toke
 Environment=AIOS_GITOPS_BARE_REPO=/data/aios-gitops/runtime/gitea-bin/data/repositories/aios/aios-v23.git
 Environment=AIOS_GITOPS_HOST=127.0.0.1
 Environment=AIOS_GITOPS_PORT=18890
 EnvironmentFile=-/etc/aios-gitops/orchestrator-callback.env
 ExecStart=/usr/bin/python3 /data/aios-gitops/workspace/active-runtime/app/aios_chatgpt_actions_api.py
 Restart=always
 RestartSec=3
@@ -1,137 +0,0 @@
 # AIOS GPT + one-executor automation quorum
 ## Authority and release boundary
 The Gateway is the only authoritative state writer. It must hold the SQLite writer lease before changing task state. A model response, service process, configured credential, or agent presence does not establish health. Release remains paused until all live acceptance rows are backed by logs and exit codes.
 Minimum quorum:
 ```text
 central_gpt=HEALTHY AND (openclaw=HEALTHY OR codex=HEALTHY)
 ```
 `UNKNOWN` never counts as healthy. If one executor is proven healthy, an unhealthy or unknown standby is P3 telemetry only and cannot block the active gate. If no executor is proven healthy, the controller blocks. Central GPT `UNKNOWN` or `UNHEALTHY` always blocks.
 ## Live health receipts
 The controller reads three root-owned metadata-only JSON receipts from `/run/aios-health`. Files must be regular files, must not be symlinks, and neither files nor parent directories may be group/world writable. The controller rejects fields named `secret`, `token`, `password`, `api_key`, `authorization`, `credential`, `private_key`, `environment`, `env`, `argv`, `stdout`, or `stderr` at any depth.
 A receipt claiming `HEALTHY` must contain:
 ```json
 {
  "subject": "codex",
  "health": "HEALTHY",
  "live_probe_succeeded": true,
  "observed_at_epoch": 1784347200
 }
 ```
 The central GPT receipt additionally requires either `workspace_agent.validated_live=true` or `responses_api.validated_live=true`. Responses fallback is allowed only after a real Responses decision probe succeeded; credential/configuration presence is insufficient.
 The controller and probes use only the fixed loopback endpoint
 `http://127.0.0.1:18920/v1/orchestrator/events`. The bearer is read from
 `/etc/aios-orchestrator-gateway/inbound.token`; the file must be absolute,
 root-owned, regular, non-symlinked, no wider than `0640`, nonempty, and contain
 no CR/LF. It is never accepted through argv, environment, logs, receipts, or
 Git. HTTP `401`, `403`, `409`, and `422` are terminal fail-closed results for a
 probe attempt.
 The authenticated metadata query
 `GET /v1/orchestrator/events/<event_id>` returns only the event ID, public
 status, selected central channel, whether the decision was applied, and the
 resulting state version. It never returns the submitted event, model output,
 credential, or work-unit instruction.
 ## Durable behavior
 SQLite persists ingress event digests, optimistic task state versions, leases, checkpoints, lifecycle events, effect IDs, outbox messages, blocker counts, observed Git refs, and project summaries. Reusing an event/effect/outbox ID with different content fails closed. An expired lease may be reclaimed; a live lease cannot be stolen.
 On active-executor failure, a healthy standby receives the same `contract_commit`, `event_id`, state version, and checkpoint. The former executor is recorded as `INTERRUPTED_AGENT_FAILOVER`. Unique effect IDs prevent replay of completed side effects.
 The Gateway persists these events even when OpenClaw is offline:
 ```text
 TASK_DISPATCHED
 TASK_STARTED
 TASK_ENDED
 DECISION_APPLIED
 PROJECT_SUMMARY
 PROJECT_COMPLETED
 ```
 OpenClaw mirroring is optional and uses the durable outbox with bounded retry and dead-letter. HTTP ingress failure spools a metadata-only file for polling; the configured local Git refs are also polled and changes missed by webhooks become idempotent events. Repeated blockers beyond the configured threshold enter `DEAD_LETTER`.
 ## Install, verify, and rollback
 Installation is versioned and does not overwrite an existing release:
 ```bash
 sudo scripts/install_automation_quorum_controller.sh "$PWD" 20260718.1
 sudo systemctl enable --now aios-automation-quorum-controller.timer
 sudo systemctl enable --now aios-central-gpt-health-probe.timer
 sudo systemctl enable --now aios-codex-executor-health-probe.timer
 ```
 Run deterministic verification before enablement:
 ```bash
 scripts/verify_automation_quorum_controller.sh "$PWD"
 ```
 The two root probes are fixed, bounded, side-effect-free programs. They write
 receipts atomically with mode `0600` below the root-owned `0750`
 `/run/aios-health`. The Codex probe may write only one `acceptance_probe`
 command into the controlled watcher queue; it cannot accept arbitrary command
 text from configuration.
 Run the strict two-cycle closure only after both services are installed from an
 immutable committed release:
 ```bash
 sudo /opt/aios-automation-quorum/current/bin/a0-unattended-loop-probe \
  --config /etc/aios-automation-quorum/health-probes.json
 ```
 Each cycle requires a live central decision that dispatches the named Codex
 acceptance work unit through the Gateway. The verifier then requires exactly
 one claim/start/execute/end/post marker, both lifecycle events in the Gateway,
 and a persisted `PROJECT_SUMMARY`. Directly queuing a command cannot satisfy
 this two-cycle closure.
 Rollback accepts only an existing non-symlink directory below `/opt/aios-automation-quorum/releases`. Repeating rollback to the current target is a no-op:
 ```bash
 sudo scripts/rollback_automation_quorum_controller.sh \
  /opt/aios-automation-quorum/releases/20260718.0
 ```
 ## Live acceptance matrix
 Each row requires captured logs, timestamps, exit codes, committed OIDs, Gitleaks output, and where applicable live Gateway DB evidence. Static unit tests are necessary but cannot close the gate.
 1. GPT + OpenClaw healthy, Codex stopped: two unattended cycles.
 2. GPT + Codex healthy, OpenClaw stopped: two unattended cycles.
 3. Both executors healthy: OpenClaw gets repetitive infrastructure; Codex gets complex code/security.
 4. Active stops mid-task: standby continues from checkpoint with no duplicate effect.
 5. Standby alone stops: active flow continues and no repair gate is created.
 6. GPT stops: execution blocks even with two healthy executors.
 7. GPT healthy and both executors stop: `BLOCKED_NO_HEALTHY_EXECUTOR`.
 8. `UNKNOWN` is never counted as healthy.
 9. HTTP ingress failure recovers through file polling.
 10. Missed webhook recovers through Git ref polling.
 11. Gateway restart restores events, leases, and outbox.
 12. Repeated event ID does not create a second dispatch; changed payload dead-letters.
 13. Repeated blocker beyond threshold dead-letters.
 14. Workspace Agent failure falls back only with a live-validated Responses channel.
 15. OpenClaw offline: Codex lifecycle and project summary remain complete in Gateway.
 16. Codex offline: OpenClaw lifecycle and project summary remain complete in Gateway.
 Until all rows pass against the live service, retain:
 ```text
 AUTOMATION_CLOSED=NOT_STARTED
 BASELINE=V1R4C6
 RELEASE=paused
 PRODUCTION_RELEASE=NOT_STARTED
 ```
@@ -0,0 +1,243 @@
 # AIOS V2 用户需求规格
 - 文档版本：`V2R1C0-candidate`
 - 状态：候选版本需求基线
 - 适用范围：AIOS Control Plane、Agent Router、OpenClaw Adapter、Hermes Runtime、Web Console、GitOps、审计与部署
 ## 1. 背景与目标
 现有系统已经具备 OpenClaw、Hermes、Agent Router、GitOps、局部 WebUI、审计和多租户候选能力。AIOS V2 的目标不是推倒重写，而是将这些分散能力收敛到统一、可审计、可 Web 化、可多租户隔离的控制面。
 AIOS V2 必须成为唯一控制面，负责租户、身份、Agent、策略、任务生命周期、审批、审计、Runtime 目录和执行证据。OpenClaw 与 Hermes 只承担执行职责，不得持有第二套控制面状态。
 ## 2. 用户角色
 | 角色 | 主要诉求 |
 |---|---|
 | 平台管理员 | 管理租户、运行时、发布、GitOps 状态和全局策略 |
 | 租户管理员 | 管理本租户 Agent、任务、审批、审计和成员权限 |
 | 操作员 | 创建任务、发起执行、取消任务、查看运行状态 |
 | 审批人 | 审批高风险任务，查看策略判定依据 |
 | 审计员 | 查询本租户完整审计链和执行证据 |
 | Runtime Worker | 领取已批准任务、心跳、返回执行结果 |
 | 开发与运维人员 | 通过 GitOps 提交、测试、部署、回滚和修复 Bug |
 ## 3. 核心用户需求
 ### UR-001 唯一控制面
 所有任务创建、审批、取消、状态变化、策略判定和审计必须由 AIOS Control Plane 统一处理。
 验收：
 - Web Console、API、Codex、OpenClaw 和 Hermes 均不得绕过 AIOS 创建或修改任务生命周期。
 - OpenClaw/Hermes 的公开接口仅限执行、状态、取消、心跳和结果回传。
 ### UR-002 复用现有能力
 系统不得重写已经稳定的 OpenClaw 执行能力、Hermes Runtime、Agent Router 或 GitOps 基础设施。
 验收：
 - 通过 Adapter/Boundary 接入现有能力。
 - 替换控制逻辑时保留原执行接口和可迁移数据。
 ### UR-003 OpenClaw 与 Hermes 解耦
 OpenClaw 与 Hermes 必须从策略、租户、审批和任务生命周期中逐步解耦。
 验收：
 - OpenClaw 仅接收已签名、已批准、带租户上下文的命令。
 - Hermes 仅接收已批准的租户分区 Job。
 - 二者均不得直接写 Control Plane 权威状态。
 ### UR-004 全面 Web 化
 所有面向用户和运维的能力必须通过 Web Console 或稳定 API 访问。
 Web Console 至少展示：
 - Tenant、Agent、Task、Approval、Runtime、Audit、Execution Evidence、GitOps Status。
 - 所有写操作由 AIOS API 完成，浏览器不得直接调用 OpenClaw/Hermes。
 ### UR-005 多租户隔离
 所有模块必须支持租户隔离，且默认拒绝跨租户访问。
 隔离范围：
 - API 与会话
 - Tenant/Agent/Task 数据
 - OpenClaw 命令
 - Hermes Queue、Cell、Profile、Worker Lease
 - Memory、Cache、Artifact、SecretRef
 - Audit、Execution Evidence、日志和指标
 验收：
 - 资源记录的 `tenant_id` 不可为空。
 - 跨租户访问返回拒绝或“资源不存在”，不得泄露存在性。
 - 必须有跨租户负向测试。
 ### UR-006 权威任务生命周期
 任务必须遵循统一状态机：
 ```text
 CREATED -> APPROVED -> RUNNING -> COMPLETED
    |          |          |
    +----------+----------+-> CANCELLED
                           +-> FAILED
 ```
 验收：
 - 禁止越级变化，例如 `CREATED -> COMPLETED`。
 - 每次变化包含 revision、principal、trace、时间和审计事件。
 ### UR-007 策略与审批
 所有有副作用任务必须经过策略判定；高风险任务必须支持人工审批。
 验收：
 - Policy Decision 包含 allow/deny、reason 和 policy version。
 - 拒绝事件进入审计。
 - Runtime 不得自行放宽策略。
 ### UR-008 幂等与并发安全
 任务和 Runtime 执行必须支持幂等与乐观并发控制。
 验收：
 - 相同租户与幂等键、相同内容返回同一任务/执行。
 - 相同幂等键不同内容返回冲突。
 - revision 或 branch lease 冲突不得静默覆盖。
 ### UR-009 可观测与执行证据
 平台必须能够确认 Codex、OpenClaw、Hermes 等执行器是否领取任务、当前进度、最近提交、测试结果和完成状态。
 验收：
 - 存在 claim、heartbeat、progress、result 和 evidence 记录。
 - Web/API 可按租户查询。
 - 执行证据不得包含明文密钥。
 ### UR-010 GitOps 唯一变更流程
 所有代码、配置、文档和部署资产修改必须进入非 `main` 分支，通过 commit、测试、审查和受控提升。
 验收：
 - Controller worktree 必须 detached 且 clean。
 - 本地 bare repo 与 Gitea 目标分支 OID 一致。
 - 不允许直接 push `main`、直接创建生产 Tag 或绕过审批部署。
 ### UR-011 持久化与恢复
 租户、Agent、Task、Audit、Idempotency 和 Execution Evidence 必须在服务重启后恢复。
 验收：
 - 候选版支持 SQLite WAL 原子快照。
 - 有 restart recovery 测试。
 - 生产版需进一步提供迁移、备份、恢复、HA 和容量证据。
 ### UR-012 审计与合规
 所有控制与执行动作必须生成不可混租户的审计事件。
 至少记录：
 - tenant、principal、trace、action、resource、outcome、policy version、timestamp。
 - OpenClaw/Hermes Runtime 生命周期。
 - GitOps、部署和回滚动作。
 ### UR-013 密钥安全
 Git、日志、Web、任务 payload 和审计记录中不得出现明文 API Key、密码、Token 或私钥。
 验收：
 - Git 只存 `secret_ref`、`credential_ref`、`key_id` 或受保护文件路径。
 - Secret 文件权限为 `0600` 或更严格。
 - Secret Scanner 默认拒绝可疑提交。
 ### UR-014 部署与回滚
 系统必须提供 systemd、Docker/Compose、健康检查和受保护回滚入口。
 验收：
 - 服务以非 root 权限运行。
 - 状态目录独立持久化。
 - 回滚前自动备份状态。
 - 默认只回滚代码，状态恢复需要显式授权。
 ### UR-015 文档与代码同步
 任何影响用户行为、架构边界、API、数据模型、状态机、安全策略、部署或运维方式的变更，必须在同一分支同步更新 `docs/aios-v2/`。
 验收：
 - 架构/功能代码变化但无文档变化时，文档同步门禁失败。
 - Traceability Matrix 必须指向当前代码和测试。
 - 新架构决策必须增加或更新 ADR。
 ## 4. 非功能需求
 ### NFR-001 安全
 - 默认拒绝、最小权限、租户隔离、身份可信链、SecretRef、审计。
 ### NFR-002 可靠性
 - 幂等、revision、worker lease、heartbeat、重试边界、原子持久化、健康检查和回滚。
 ### NFR-003 可维护性
 - 模块边界稳定，Control Plane 与 Runtime 解耦；代码、文档、测试和部署可追踪。
 ### NFR-004 可扩展性
 - Store、Transport、Policy、EventPublisher、Runtime Adapter 均通过接口扩展。
 ### NFR-005 可运维性
 - Web/API 状态、systemd/Docker、健康检查、日志、GitOps 状态、候选报告。
 ### NFR-006 性能候选指标
 候选版上线前至少验证：
 - API P95、任务调度延迟、队列等待、Worker 心跳、SQLite 锁等待、审计写入时间。
 - 具体 SLO 在预发布基准测试后固化，禁止无证据承诺。
 ## 5. 本候选版本范围
 ### 已纳入代码
 - Control Plane、持久化、Gateway、Orchestrator、OpenClaw Adapter、Hermes Boundary、Execution Tracking、Web Console、GitOps Status、部署和回滚。
 ### 尚需环境验证
 - 精确测试执行证据。
 - OpenClaw 真实 Runtime 联调。
 - 生产数据库/队列替换或容量验证。
 - 预发布性能、安全、重启和回滚演练。
 - Owner 对 merge、Tag 和生产部署的批准。
 ## 6. 用户验收场景
 1. 平台管理员创建租户并注册 OpenClaw/Hermes Agent。
 2. 操作员创建任务，审批人批准。
 3. AIOS Orchestrator 选择 Runtime，并创建签名命令或租户分区 Job。
 4. Runtime 执行并返回结果。
 5. AIOS 更新 Task、Audit 和 Execution Evidence。
 6. 租户用户在 Web Console 查看完整链路。
 7. 跨租户访问被拒绝。
 8. 服务重启后状态恢复。
 9. 候选部署健康检查通过，回滚演练成功。
@@ -0,0 +1,461 @@
 # AIOS V2 目标架构
 - 文档版本：`V2R1C0-candidate`
 - 架构状态：目标架构已实现为候选代码，待预发布验证
 - 适用范围：Control Plane、Agent Router、OpenClaw、Hermes、Web Console、GitOps、审计、部署与运维
 ## 1. 架构目标
 AIOS V2 的核心目标是把分散在 OpenClaw、Hermes、Agent Router、局部 WebUI 和 GitOps 中的控制能力收敛为一个权威控制面，同时保留现有执行能力，避免推倒重写。
 目标状态：
 ```text
 用户 / Web Console / Public API
                 |
                 v
          AIOS Control Plane
 Tenant | Identity | Agent | Task | Policy | Approval | Audit
                 |
                 v
      Execution Orchestrator / Agent Router
             /                      \
            v                        v
   OpenClaw Adapter            Hermes Boundary
            |                        |
            v                        v
   OpenClaw Runtime            Hermes Workers
             \                      /
              v                    v
       Result Reconciliation + Execution Evidence
 ```
 设计结论：
 - AIOS 是唯一控制面。
 - Agent Router 是调度平面，不是第二控制面。
 - OpenClaw 是执行适配与 Runtime。
 - Hermes 是队列、Worker、Profile 与执行 Runtime。
 - Web Console 只调用 AIOS API。
 - GitOps 是代码、配置、文档和发布变更的唯一入口。
 ## 2. 从现有架构到目标架构
 ### 2.1 现有能力
 现有系统已经具备：
 - OpenClaw 执行与 Fleet 能力。
 - Hermes Worker、Profile、队列和工作区能力。
 - Agent Router 路由经验和部分调度接口。
 - GitOps 分支、提交、审批、回滚和审计能力。
 - Tenant Context、Tenant Router、Memory Scope、SecretRef、Message Ledger 和审计候选实现。
 - 局部 WebUI、Gateway 和自动任务领取能力。
 ### 2.2 现有问题
 - 控制逻辑分散，可能存在多个任务事实源。
 - OpenClaw/Hermes 内部仍可能保存租户、策略或任务状态。
 - 候选能力分散在多个分支，主线缺少权威实现。
 - WebUI 与 API 入口不统一。
 - 分支名称、代码、测试和部署证据之间缺少映射。
 - 多租户隔离多为逻辑约束，尚缺全链路生产证据。
 ### 2.3 演进原则
 - 先加边界和适配器，再迁移控制权。
 - 先复用接口，再替换内部实现。
 - 先在候选分支验证，再受控提升。
 - 不允许同时存在两套权威任务生命周期。
 - 不以“大分支存在”代替验收证据。
 ## 3. 逻辑分层
 ### 3.1 Web 与 API 层
 职责：
 - 提供 Tenant、Agent、Task、Approval、Runtime、Audit、Execution Evidence 和 GitOps 页面/API。
 - 依赖可信身份代理注入 `X-AIOS-*` 身份头。
 - 不允许浏览器直接访问 OpenClaw 或 Hermes。
 - 不允许客户端自行选择或覆盖 `tenant_id`。
 候选实现：
 - `active-runtime/app/aios_v2/app.py`
 - `active-runtime/app/aios_v2/gateway.py`
 - `active-runtime/app/aios_v2/console.py`
 ### 3.2 AIOS Control Plane
 权威职责：
 - Tenant Registry
 - Agent Registry
 - Task Lifecycle
 - Policy Decision
 - Approval
 - Audit Event
 - Runtime Inventory
 - Execution Evidence
 - Deployment/GitOps Status
 不负责：
 - OpenClaw 内部执行细节。
 - Hermes Worker 内部执行细节。
 - 直接持有 Runtime 凭据明文。
 候选实现：
 - `active-runtime/app/aios_v2/control_plane.py`
 - `active-runtime/app/aios_v2/persistence.py`
 - `active-runtime/app/aios_v2/execution_tracking.py`
 - `active-runtime/app/aios_v2/gitops_status.py`
 ### 3.3 Execution Orchestrator / Agent Router
 职责：
 - 只接受 `APPROVED` 任务。
 - 根据 Agent Registry 和 Policy Result 选择 Runtime。
 - 生成 OpenClawCommand 或 HermesJob。
 - 传播 tenant、task、trace、idempotency 和 policy version。
 - 将 Runtime Result 回写 AIOS Task。
 不负责：
 - 创建租户。
 - 修改策略。
 - 绕过审批。
 - 保存第二套任务状态机。
 候选实现：
 - `active-runtime/app/aios_v2/orchestrator.py`
 ### 3.4 OpenClaw Adapter
 职责：
 - 验证签名命令。
 - 执行、查询、取消、健康检查和回调。
 - 支持幂等、超时和租户隔离。
 - 把 Runtime 生命周期转换为 AIOS 审计事件。
 不负责：
 - Tenant、Policy、Approval、Task Lifecycle。
 - 向浏览器暴露公网控制接口。
 候选实现：
 - `active-runtime/app/aios_openclaw/contract.py`
 - `active-runtime/app/aios_openclaw/adapter.py`
 - `active-runtime/app/aios_openclaw/transport.py`
 - `active-runtime/app/aios_openclaw/health.py`
 - `active-runtime/app/aios_openclaw/callback.py`
 ### 3.5 Hermes Runtime Boundary
 职责：
 - Tenant/Cell/Profile 分区队列。
 - Worker claim、start、heartbeat、complete、fail、cancel。
 - Lease 超时和自动回队列。
 - Result 回传和 Runtime Health。
 不负责：
 - Policy、Approval、Tenant Registry、权威 Task Lifecycle。
 候选实现：
 - `active-runtime/app/aios_hermes/contract.py`
 - `active-runtime/app/aios_hermes/adapter.py`
 - `active-runtime/app/aios_hermes/worker.py`
 ### 3.6 Persistence
 候选实现使用 SQLite WAL 原子快照，持久化：
 - Tenant
 - Agent
 - Task
 - Audit
 - Idempotency
 - Execution Evidence 文件
 生产演进：
 - Store 接口保持稳定。
 - 可替换为 PostgreSQL、对象存储或分布式事件存储。
 - 生产替换必须提供迁移、备份、恢复、HA、容量和租户隔离证据。
 ## 4. 多租户架构
 ### 4.1 权威租户上下文
 每次请求和执行至少传播：
 ```text
 tenant_id
 principal_id
 session_id
 trace_id
 task_id
 policy_version
 ```
 Runtime 还传播：
 ```text
 command_id / job_id
 cell_id
 profile_id
 idempotency_key
 worker_id
 ```
 ### 4.2 隔离层级
 | 层级 | 隔离要求 |
 |---|---|
 | HTTP/API | 可信身份头、租户不可由浏览器覆盖 |
 | Control Plane | 所有资源带 `tenant_id`，跨租户默认拒绝 |
 | OpenClaw | 签名命令绑定 tenant/task/trace |
 | Hermes | tenant/cell/profile 三层分区 |
 | Storage | 候选为显式租户字段；生产需 RLS 或物理隔离策略 |
 | Evidence/Audit | 仅按当前租户查询 |
 | Cache/Memory | key 必须包含 tenant namespace |
 | Secret | 只保存引用，不保存明文 |
 | Logs/Metrics | 必须包含 tenant/task/trace 标签，并限制查询范围 |
 ### 4.3 失败行为
 - 跨租户资源访问返回拒绝或“未找到”。
 - 不泄露其他租户资源存在性。
 - 缺少 tenant context 时 fail closed。
 - Runtime 返回的 tenant/task/trace 不匹配时拒绝回写。
 ## 5. 任务与执行数据流
 ### 5.1 任务创建与审批
 ```text
 User/API
   |
   v
 Trusted Identity
   |
   v
 AIOS Gateway
   |
   v
 Policy Gate
   |
   v
 Task CREATED
   |
   v
 Approval
   |
   v
 Task APPROVED
 ```
 ### 5.2 OpenClaw 执行
 ```text
 APPROVED Task
   |
   v
 Execution Orchestrator
   |
   v
 Signed OpenClawCommand
   |
   v
 OpenClaw Adapter
   |
   v
 Runtime Result
   |
   v
 Task Reconciliation + Audit + Evidence
 ```
 ### 5.3 Hermes 执行
 ```text
 APPROVED Task
   |
   v
 Execution Orchestrator
   |
   v
 Tenant/Cell/Profile HermesJob
   |
   v
 Worker Claim + Lease + Heartbeat
   |
   v
 HermesResult
   |
   v
 Task Reconciliation + Audit + Evidence
 ```
 ## 6. 安全架构
 ### 6.1 身份边界
 - Public API 位于可信反向代理之后。
 - 反向代理完成登录、会话、MFA 和身份头注入。
 - 候选应用拒绝未认证身份断言。
 - 生产不得允许客户端直接伪造 `X-AIOS-*`。
 ### 6.2 Runtime 命令完整性
 OpenClawCommand 使用规范化序列化和 HMAC-SHA256：
 - 签名覆盖 tenant、task、trace、action、payload、idempotency 和时间。
 - 支持 key ID、到期时间和时钟偏差。
 - Runtime 密钥通过受保护文件或外部 Secret Provider 解析。
 ### 6.3 Secret 边界
 Git 允许：
 ```text
 secret_ref
 credential_ref
 key_id
 受保护文件路径
 ```
 Git 禁止：
 ```text
 API Key
 密码
 Token
 私钥
 签名材料
 ```
 ### 6.4 Web 安全
 候选应用设置：
 - no-store
 - nosniff
 - DENY frame
 - no-referrer
 - CSP
 - 请求体大小限制
 生产还需验证：
 - CSRF/CORS
 - Rate Limit
 - Session Timeout
 - CSP Nonce/Hash
 - WAF 和反向代理访问策略
 ## 7. 可靠性架构
 ### 7.1 幂等
 - Control Plane：`tenant_id + idempotency_key + request_digest`。
 - OpenClaw：`tenant_id + idempotency_key + signature`。
 - Hermes：`tenant_id + idempotency_key + job_digest`。
 ### 7.2 并发
 - Task 使用 revision。
 - Hermes 使用 revision + worker lease。
 - GitOps 使用 detached controller + branch lease。
 - 持久化使用 SQLite `BEGIN IMMEDIATE` 和原子快照。
 ### 7.3 恢复
 - Controller restart：detached guard。
 - AIOS restart：SQLite restore。
 - Worker crash：lease 到期后 Job 回到 QUEUED。
 - Release failure：健康检查失败自动恢复 previous release。
 ## 8. 可观测架构
 统一关联键：
 ```text
 tenant_id
 principal_id
 trace_id
 task_id
 agent_id
 command_id / job_id
 worker_id
 commit
 ```
 Web Console 展示：
 - Tasks
 - Agents
 - Runtime Health
 - Audit
 - Execution Evidence
 - GitOps Status
 生产还需接入：
 - 结构化日志
 - Metrics
 - Distributed Tracing
 - Alerting
 - SLO Dashboard
 ## 9. 部署架构
 支持两种候选部署：
 ### systemd
 - 非 root `aios` 用户。
 - `ProtectSystem=strict`、`NoNewPrivileges` 等安全选项。
 - 状态目录 `/var/lib/aios-v2`。
 ### Docker/Compose
 - 非 root UID/GID。
 - Read-only root filesystem。
 - Drop all capabilities。
 - State volume。
 - HTTP healthcheck。
 ## 10. 架构约束
 以下行为视为架构违规：
 1. OpenClaw/Hermes 直接创建或批准 AIOS Task。
 2. Web Console 直接调用 Runtime。
 3. 新模块不携带 `tenant_id`。
 4. 新增第二套任务状态机或租户事实源。
 5. 在 Git 中提交明文凭据。
 6. 绕过 GitOps 修改生产代码/配置。
 7. 架构或功能变化未同步更新 `docs/aios-v2/`。
 8. 未通过测试和预发布门禁直接提升候选版本。
 ## 11. 目标架构完成判定
 目标架构只有在以下证据齐全时才视为完成：
 - 文档与代码映射完整。
 - Candidate tests 对精确 commit 执行通过。
 - OpenClaw/Hermes 实际 Runtime 联调通过。
 - 多租户负向测试通过。
 - 重启恢复和回滚演练通过。
 - GitOps 本地/Gitea OID 一致。
 - Owner 批准 merge、Tag 和生产部署。
@@ -0,0 +1,313 @@
 # AIOS V2 迁移与交付策略
 - 文档版本：`V2R1C0-candidate`
 - 交付原则：增量迁移、能力复用、控制权收敛、GitOps 唯一变更入口
 ## 1. 迁移目标
 AIOS V2 不通过一次性重写替换现有系统，而是在保留 OpenClaw、Hermes、Agent Router、GitOps 和既有运行能力的前提下，将租户、策略、任务、审批、审计和发布治理收敛到 AIOS Control Plane。
 迁移完成的标志不是“新代码存在”，而是：
 - AIOS 成为唯一权威控制面；
 - OpenClaw/Hermes 不再保存第二套控制状态；
 - 所有控制功能 Web/API 化；
 - 所有模块通过租户隔离、测试和预发布门禁；
 - 本地 Git、Gitea、测试、部署和回滚证据可追踪。
 ## 2. 为什么采用增量迁移
 ### 2.1 避免能力回退
 OpenClaw、Hermes 和 Agent Router 已经包含大量运行经验、边界处理和执行能力。整体重写会扩大风险面，并导致业务能力、操作流程和历史数据回退。
 ### 2.2 保持可回滚
 每个迁移步骤必须：
 - 独立分支；
 - 小范围提交；
 - 独立测试；
 - 可单独回滚；
 - 不要求立即关闭旧执行路径。
 ### 2.3 先收敛控制权，再优化内部实现
 迁移优先级：
 ```text
 唯一入口
  -> 权威租户/任务/策略
  -> Runtime Adapter
  -> 数据层生产化
  -> Web 与可观测
  -> 性能与 HA
 ```
 ## 3. 分阶段迁移路线
 ### Phase 0：GitOps 与架构事实源
 目标：建立可靠变更链和权威设计基线。
 已完成候选能力：
 - Controller detached guard；
 - 避免 fetch 到已检出本地分支；
 - Gitea SSH 镜像；
 - `docs/aios-v2/` 架构事实源；
 - 候选分支、部署和回滚资产。
 退出条件：
 - Controller detached 且 clean；
 - 本地/Gitea OID 一致；
 - 架构与需求文档纳入版本。
 ### Phase 1：Control Plane 权威化
 目标：将 Tenant、Agent、Task、Policy、Approval 和 Audit 收敛到 AIOS。
 候选实现：
 - `aios_v2/control_plane.py`
 - `aios_v2/persistence.py`
 - `aios_v2/gateway.py`
 - `aios_v2/app.py`
 退出条件：
 - 任务状态机统一；
 - 幂等、revision、审计和跨租户拒绝通过测试；
 - 服务重启后权威状态恢复。
 ### Phase 2：OpenClaw 解耦
 目标：OpenClaw 从控制逻辑转为执行适配器。
 步骤：
 1. 定义签名命令合同；
 2. 引入 Adapter；
 3. 将直接调用迁移到 AIOS Orchestrator；
 4. Runtime 只返回执行结果和健康状态；
 5. 关闭或网络隔离旧公共控制入口。
 退出条件：
 - 未批准任务无法生成有效命令；
 - 命令绑定 tenant/task/trace/idempotency；
 - 旧控制接口冻结并记录替代路径。
 ### Phase 3：Hermes 解耦
 目标：Hermes 只承担租户分区队列和 Worker Runtime。
 步骤：
 1. 定义 HermesJob/HermesResult；
 2. 强制 tenant/cell/profile 分区；
 3. 引入 claim/lease/heartbeat；
 4. 将结果统一回写 AIOS；
 5. 移除 Hermes 内部策略和任务事实源。
 退出条件：
 - Worker 无权创建 AIOS Task；
 - 租约过期后 Job 可恢复；
 - 跨租户 claim/read/result 被拒绝。
 ### Phase 4：Web 与运维统一
 目标：所有用户与运维功能通过 AIOS Web/API。
 范围：
 - Tenant、Agent、Task、Runtime、Audit、Execution Evidence、GitOps；
 - systemd/Docker/Compose；
 - healthcheck、rollback、release report。
 退出条件：
 - 浏览器不直接调用 Runtime；
 - 生产身份代理和权限模型通过验证；
 - 部署、重启和回滚演练通过。
 ### Phase 5：生产数据层与 HA
 目标：将候选 SQLite/内存运行状态替换或强化为生产级后端。
 要求：
 - 数据迁移；
 - RLS 或物理隔离；
 - 备份恢复；
 - HA 与容量；
 - 多实例并发；
 - 灾难恢复目标。
 该阶段不得改变上层合同，避免 Runtime 和 Web 再次耦合存储实现。
 ## 4. 分支与提交策略
 ### 4.1 分支类型
 | 类型 | 用途 |
 |---|---|
 | `task/*` | 可执行任务合同与主体实现 |
 | `patch/*` | 独立修复或窄范围增强 |
 | `integration/*` | 多模块候选收敛 |
 | `result/*` | 自动执行结果与证据 |
 | `release/*` | 受控发布候选 |
 ### 4.2 规则
 - 禁止直接修改 `main`；
 - 一个提交应可独立审阅和回滚；
 - 大文件和大范围变更应拆分；
 - 所有行为变化必须同步文档和测试；
 - 生产 Tag、部署和回滚需要 Owner 批准。
 ## 5. GitOps 交付流程
 ```text
 需求/架构更新
      |
      v
 任务分支 + 文档 + 测试
      |
      v
 代码提交 + Secret Gate
      |
      v
 本地 bare repo 同步
      |
      v
 Gitea 镜像与 OID 校验
      |
      v
 Candidate 集成
      |
      v
 自动测试与预发布
      |
      v
 Owner 审批
      |
      v
 Merge / Tag / Production Deploy
 ```
 ### 5.1 同步状态
 GitOps 返回必须区分：
 - local sync；
 - Gitea sync；
 - branch OID；
 - workspace mode；
 - dirty state。
 仅返回 `common_git_dir synced` 不足以证明 Web 远端可见。
 ## 6. 发布门禁
 ### Gate A：设计完整性
 - 用户需求、目标架构、模块契约、ADR 和追踪矩阵同步；
 - 无冲突标记、断链或过期路径。
 ### Gate B：代码质量
 - 包可导入；
 - 静态编译通过；
 - Secret Scanner 通过；
 - 无已知越权旁路。
 ### Gate C：测试
 - 单元、集成、多租户负向、重启恢复、Runtime 边界测试通过；
 - 记录 commit、命令、exit code、pass/fail 和 artifact digest。
 ### Gate D：预发布
 - systemd 或 Compose 启动成功；
 - 健康检查通过；
 - OpenClaw/Hermes 真实联调通过；
 - 备份、重启和回滚演练通过；
 - 日志中无敏感信息。
 ### Gate E：生产批准
 - Owner 批准 merge、Tag 和生产部署；
 - 发布窗口、回滚负责人和观察指标明确；
 - 生产变更通过 GitOps 记录。
 ## 7. 回滚策略
 ### 7.1 代码回滚
 - 使用 `/opt/aios/current` 原子 symlink 切换；
 - 回滚前备份状态；
 - 健康检查失败自动恢复 previous release。
 ### 7.2 状态回滚
 默认不自动回滚状态，避免覆盖候选运行期间新增的 Task、Audit 和 Evidence。
 状态恢复只有在：
 - 明确确认 schema 不兼容；
 - 数据备份可验证；
 - Owner 显式授权；
 时启用。
 ### 7.3 Runtime 回滚
 OpenClaw/Hermes Adapter 采用稳定合同。若候选执行层失败，可恢复旧 Runtime 版本，但不得恢复旧控制权旁路。
 ## 8. Bug 修复流程
 生产或预发布 Bug：
 1. 创建 `patch/bugfix/*`；
 2. 添加可复现测试；
 3. 修复代码；
 4. 更新相关需求/架构/契约/ADR；
 5. 运行候选验收；
 6. 进入新的 release candidate；
 7. Owner 批准部署。
 禁止在生产目录直接编辑文件后不回写 Git。
 ## 9. 文档同步规则
 以下变化必须更新文档：
 - 用户行为或页面；
 - API 路径、Header、请求/响应；
 - 数据模型、状态机、策略；
 - OpenClaw/Hermes/Router 边界；
 - 多租户与安全模型；
 - 部署、健康检查、回滚；
 - 测试或发布门禁。
 对应关系：
 - 需求变化：更新 `01-USER-REQUIREMENTS.md`；
 - 架构变化：更新 `02-TARGET-ARCHITECTURE.md` 和 ADR；
 - 迁移/发布变化：更新本文件；
 - 合同变化：更新 `04-MODULE-CONTRACTS.md`；
 - 代码/测试路径变化：更新 `05-TRACEABILITY-MATRIX.md`。
 ## 10. 当前候选后续步骤
 1. 完成文档同步门禁和 Acceptance Runner；
 2. 在精确 Candidate commit 执行测试；
 3. 推送 Gitea 并校验 OID；
 4. 部署隔离预发布环境；
 5. 联调真实 OpenClaw/Hermes；
 6. 执行重启、性能、安全和回滚演练；
 7. 生成最终 evidence；
 8. 请求 Owner 的 merge/Tag/production approval。
@@ -0,0 +1,502 @@
 # AIOS V2 模块契约
 - 文档版本：`V2R1C0-candidate`
 - 状态：候选实现契约基线
 - 原则：Control Plane 权威、Runtime 执行、全链路租户隔离、所有状态可审计
 ## 1. 通用上下文契约
 所有控制与执行请求必须携带以下上下文：
 | 字段 | 必填 | 说明 |
 |---|---:|---|
 | `tenant_id` | 是 | 权威租户标识 |
 | `principal_id` | 控制请求是 | 已认证主体 |
 | `trace_id` | 是 | 全链路追踪 |
 | `task_id` | 执行阶段是 | AIOS 权威任务 |
 | `session_id` | Web/交互请求建议 | 用户会话 |
 | `policy_version` | 控制请求是 | 策略版本 |
 | `idempotency_key` | 创建/执行是 | 幂等键 |
 规则：
 - 缺少必填字段时 fail closed。
 - 客户端不得覆盖由可信代理或 AIOS 生成的 `tenant_id`。
 - Runtime 返回的 tenant/task/trace 必须与请求完全匹配。
 - 所有持久化键、缓存键和队列分区必须包含租户命名空间。
 ## 2. Tenant Contract
 ### 2.1 TenantRecord
 ```text
 tenant_id
 display_name
 status
 metadata
 created_at
 updated_at
 ```
 约束：
 - `tenant_id` 不可为空且不可变。
 - `status` 至少支持 `active` 和受控停用状态。
 - Tenant 只由 AIOS Control Plane 创建和维护。
 - Runtime 不得创建租户。
 实现：`active-runtime/app/aios_v2/control_plane.py`
 ## 3. Agent Contract
 ### 3.1 AgentRecord
 ```text
 agent_id
 tenant_id
 display_name
 runtime
 capabilities
 enabled
 metadata
 created_at
 updated_at
 ```
 约束：
 - Agent 属于唯一租户。
 - `runtime` 由 Registry 决定，可取 `openclaw`、`hermes` 或后续注册 Runtime。
 - `capabilities` 是允许调度的动作集合。
 - `metadata` 可存 `cell_id`、`profile_id`、priority 等非敏感引用。
 - 不得在 Agent metadata 中保存明文凭据。
 ## 4. Task Contract
 ### 4.1 TaskRecord
 ```text
 task_id
 tenant_id
 agent_id
 action
 payload
 state
 revision
 idempotency_key
 request_digest
 result
 error
 created_by
 approved_by
 created_at
 updated_at
 ```
 ### 4.2 状态机
 ```text
 CREATED -> APPROVED -> RUNNING -> COMPLETED
    |          |          |
    +----------+----------+-> CANCELLED
                           +-> FAILED
 ```
 允许转换：
 | 当前 | 允许目标 |
 |---|---|
 | CREATED | APPROVED, CANCELLED |
 | APPROVED | RUNNING, CANCELLED |
 | RUNNING | COMPLETED, FAILED, CANCELLED |
 | COMPLETED | 无 |
 | FAILED | 无 |
 | CANCELLED | 无 |
 约束：
 - 每次转换必须校验 `revision`。
 - COMPLETED 必须带 `result`。
 - FAILED 必须带 `error`。
 - 每次转换生成 AuditEvent。
 - Runtime 不得直接写 Task Store，只能通过 Orchestrator reconcile。
 ## 5. Policy Contract
 ### 5.1 PolicyDecision
 ```text
 allowed
 reason
 policy_version
 ```
 约束：
 - Policy 默认拒绝。
 - Runtime 不得覆盖 PolicyDecision。
 - 拒绝必须审计。
 - 高风险动作必须经过 Approval。
 ## 6. Audit Contract
 ### 6.1 AuditEvent
 ```text
 event_id
 tenant_id
 principal_id
 trace_id
 action
 resource_type
 resource_id
 outcome
 policy_version
 details
 created_at
 ```
 约束：
 - 按租户查询。
 - `details` 不得包含明文凭据。
 - OpenClaw/Hermes 生命周期事件必须映射到统一 AuditEvent。
 - 生产版需增加不可变外部固化和完整性验证。
 ## 7. OpenClaw Contract
 ### 7.1 OpenClawCommand
 ```text
 contract_version
 command_id
 tenant_id
 task_id
 trace_id
 action
 payload
 idempotency_key
 created_at
 expires_at
 key_id
 signature
 ```
 签名覆盖：
 - command_id
 - tenant_id
 - task_id
 - trace_id
 - action
 - payload
 - idempotency_key
 - created_at / expires_at
 - contract_version / key_id
 约束：
 - 使用规范化 JSON 和 HMAC-SHA256。
 - 只接受已批准任务生成的命令。
 - 过期、未来时间超差、签名错误命令必须拒绝。
 - `key_id` 只引用运行时材料，不包含材料本身。
 ### 7.2 OpenClawResult
 ```text
 command_id
 tenant_id
 task_id
 trace_id
 state
 output
 error
 started_at
 finished_at
 runtime_id
 attempts
 ```
 状态：
 ```text
 ACCEPTED
 RUNNING
 COMPLETED
 FAILED
 CANCELLED
 TIMED_OUT
 REJECTED
 ```
 约束：
 - Adapter 对外只接受终态结果用于 reconcile。
 - 身份字段不匹配时拒绝回写。
 - 结果通过 Orchestrator 更新 AIOS Task。
 ### 7.3 RuntimeHealth
 ```text
 runtime
 runtime_id
 status
 version
 heartbeat_at
 active_tasks
 workspace_isolated
 credential_ref_configured
 details
 ```
 健康策略：
 - heartbeat 不得超时。
 - workspace isolation 与 credential reference 必须符合策略。
 - 不安全 Runtime 应 quarantine，不得继续接收新任务。
 实现：`active-runtime/app/aios_openclaw/`
 ## 8. Hermes Contract
 ### 8.1 HermesJob
 ```text
 job_id
 tenant_id
 cell_id
 profile_id
 task_id
 trace_id
 action
 payload
 idempotency_key
 priority
 state
 revision
 assigned_worker_id
 lease_expires_at
 created_at
 updated_at
 ```
 状态机：
 ```text
 QUEUED -> CLAIMED -> RUNNING -> COMPLETED
   |         |          |
   +---------+----------+-> CANCELLED
                        +-> FAILED
 ```
 约束：
 - `tenant_id + cell_id + profile_id` 形成执行分区。
 - claim 只能领取同分区 Job。
 - Worker ID 不得跨分区复用。
 - Lease 到期 Job 回到 QUEUED。
 - revision 冲突不得覆盖。
 ### 8.2 HermesResult
 ```text
 job_id
 tenant_id
 task_id
 trace_id
 state
 worker_id
 output
 error
 started_at
 finished_at
 ```
 约束：
 - COMPLETED 必须有 output。
 - FAILED 必须有 error。
 - 只允许 AIOS Orchestrator reconcile。
 ### 8.3 HermesHealth
 ```text
 runtime_id
 status
 version
 heartbeat_at
 active_workers
 active_jobs
 queued_jobs
 tenant_partitions
 profile_isolation_enforced
 details
 ```
 实现：`active-runtime/app/aios_hermes/`
 ## 9. Execution Evidence Contract
 ### 9.1 ExecutionEvidence
 ```text
 execution_id
 tenant_id
 task_id
 agent_id
 executor
 branch
 status
 progress_completed
 progress_total
 current_step
 last_commit
 test_status
 test_summary
 result_summary
 claimed_at
 heartbeat_at
 completed_at
 ```
 状态建议：
 ```text
 claimed
 running
 blocked
 complete
 failed
 cancelled
 ```
 约束：
 - 按租户存储和查询。
 - 每个阻塞必须记录原因、责任方和解除条件。
 - 每次结果必须记录 branch、commit、测试命令和状态。
 - 不得包含 Secret 或完整敏感 payload。
 ## 10. GitOps Status Contract
 ```text
 available
 service_status
 workspace_mode
 workspace_branch
 workspace_ref
 workspace_dirty
 candidate_branch
 release_channel
 release_version
 sync_status
 details
 ```
 约束：
 - Web 响应不得包含 remote URL、凭据或 stderr。
 - Controller 必须 `detached + clean`。
 - `sync_status` 必须区分 local 和 Gitea 镜像状态。
 ## 11. Web/API Contract
 ### 11.1 身份头
 可信代理注入：
 ```text
 X-AIOS-Authenticated
 X-AIOS-Tenant-ID
 X-AIOS-Principal-ID
 X-AIOS-Roles
 X-AIOS-Trace-ID
 X-AIOS-Session-ID
 X-AIOS-Policy-Version
 ```
 生产约束：
 - 代理必须删除客户端同名头后重新注入。
 - 应用不得直接暴露公网端口。
 ### 11.2 核心 API
 ```text
 GET  /api/v2/health
 POST /api/v2/tenants
 GET  /api/v2/tenants/current
 POST /api/v2/agents
 GET  /api/v2/agents
 POST /api/v2/tasks
 GET  /api/v2/tasks
 GET  /api/v2/tasks/{id}
 POST /api/v2/tasks/{id}/approve
 POST /api/v2/tasks/{id}/start
 POST /api/v2/tasks/{id}/complete
 POST /api/v2/tasks/{id}/fail
 POST /api/v2/tasks/{id}/cancel
 POST /api/v2/tasks/{id}/dispatch
 GET  /api/v2/audit
 GET  /api/v2/runtime
 GET  /api/v2/executions
 GET  /api/v2/gitops
 POST /api/v2/hermes/results/reconcile
 ```
 ## 12. Persistence Contract
 候选版：
 - SQLite WAL
 - `BEGIN IMMEDIATE`
 - 原子 snapshot
 - schema version
 - restart restore
 持久化范围：
 - tenants
 - agents
 - tasks
 - audit
 - idempotency
 生产替换要求：
 - 不改变 Domain/API Contract。
 - 提供 migration、backup、restore、HA、RLS/隔离和容量测试。
 ## 13. Deployment Contract
 ### systemd
 - 非 root `aios` 用户。
 - 状态目录 `/var/lib/aios-v2`。
 - 最小权限和文件系统保护。
 ### Docker
 - 非 root UID/GID。
 - read-only root filesystem。
 - drop all capabilities。
 - 独立状态 volume。
 ### Rollback
 - 回滚前备份状态。
 - 默认只回滚代码。
 - 状态恢复需显式授权。
 - 回滚后必须执行 healthcheck。
 ## 14. 契约变更规则
 任何以下变化必须在同一提交或同一分支更新本文档：
 - 字段、状态、Header、API、错误语义；
 - 租户或安全约束；
 - OpenClaw/Hermes 边界；
 - Persistence 或 Deployment Contract；
 - Evidence/Audit 格式。
 不兼容变化必须新增 ADR 和版本号，不得静默修改。
@@ -0,0 +1,113 @@
 # AIOS V2 需求—架构—代码—测试—部署追踪矩阵
 - 文档版本：`V2R1C0-candidate`
 - 目的：确保每项用户需求都能追踪到架构设计、代码实现、自动测试、部署资产和验收命令
 ## 1. 追踪规则
 每条需求至少具备：
 1. 需求编号；
 2. 架构章节；
 3. 实现路径；
 4. 测试路径；
 5. 部署/运维资产；
 6. 验收状态。
 状态定义：
 - `Implemented`：代码已进入 Candidate 分支；
 - `Test-defined`：测试代码已进入 Candidate 分支；
 - `Test-verified`：精确 commit 的测试已运行并保存证据；
 - `Pre-production verified`：预发布部署、健康检查和回滚演练已通过；
 - `Production released`：Owner 批准并完成 merge/tag/deploy。
 ## 2. 核心追踪矩阵
 | 需求 | 设计依据 | 代码实现 | 测试 | 部署/运维 | 当前状态 |
 |---|---|---|---|---|---|
 | UR-001 AIOS 唯一控制面 | `02-TARGET-ARCHITECTURE.md` §3.2/§3.3；ADR-001 | `aios_v2/control_plane.py`、`aios_v2/orchestrator.py`、`aios_v2/app.py` | `tests/aios_v2_candidate/test_candidate_flow.py` | `deploy/aios-v2/systemd/aios-v2.service` | Implemented / Test-defined |
 | UR-002 复用现有能力 | `02-TARGET-ARCHITECTURE.md` §2；`03-MIGRATION-AND-DELIVERY.md` §2 | `aios_openclaw/`、`aios_hermes/` | OpenClaw/Hermes flow tests | Candidate branch + adapter boundaries | Implemented / Test-defined |
 | UR-003 OpenClaw/Hermes 解耦 | `02-TARGET-ARCHITECTURE.md` §3.4/§3.5；ADR-002 | `aios_openclaw/adapter.py`、`aios_hermes/adapter.py` | runtime boundary tests in candidate flow suite | systemd/Docker runtime configuration | Implemented / Test-defined |
 | UR-004 全面 Web 化 | `02-TARGET-ARCHITECTURE.md` §3.1/§8 | `aios_v2/console.py`、`aios_v2/gateway.py`、`aios_v2/app.py` | API/console smoke test pending | Docker/systemd + reverse proxy requirement | Implemented / Test pending |
 | UR-005 多租户隔离 | `02-TARGET-ARCHITECTURE.md` §4；ADR-004 | `control_plane.py`、`orchestrator.py`、`aios_openclaw/*`、`aios_hermes/*`、`execution_tracking.py` | `test_cross_tenant_task_read_is_denied` | tenant-scoped state/evidence directories | Implemented / Test-defined |
 | UR-006 权威任务生命周期 | `04-MODULE-CONTRACTS.md` §4 | `control_plane.py` | OpenClaw/Hermes full-flow tests | persisted state + audit | Implemented / Test-defined |
 | UR-007 策略与审批 | `04-MODULE-CONTRACTS.md` §5 | `control_plane.py`、`gateway.py` | role/policy negative tests需补充 | reverse proxy identity + audit | Implemented / Partial test |
 | UR-008 幂等与并发安全 | `02-TARGET-ARCHITECTURE.md` §7 | `control_plane.py`、`aios_openclaw/adapter.py`、`aios_hermes/adapter.py`、`worker.py` | flow + restart tests；并发专项测试需补充 | GitOps detached guard + SQLite WAL | Implemented / Partial test |
 | UR-009 执行证据 | `04-MODULE-CONTRACTS.md` §9 | `execution_tracking.py` | `test_execution_tracking_persists_claim_heartbeat_and_result` | `/var/lib/aios-v2/executions` | Implemented / Test-defined |
 | UR-010 GitOps 唯一变更流程 | `03-MIGRATION-AND-DELIVERY.md` §4/§5；ADR-003 | GitOps Runtime、`gitops_status.py` | healthcheck 的 detached/clean 检查 | `deploy/aios-v2/healthcheck/check.sh` | Implemented / Runtime verified partially |
 | UR-011 持久化与恢复 | `02-TARGET-ARCHITECTURE.md` §3.6 | `persistence.py` | `test_durable_service_recovers_state_after_restart` | state dir + backup/rollback | Implemented / Test-defined |
 | UR-012 审计与合规 | `04-MODULE-CONTRACTS.md` §6 | `control_plane.py`、`aios_openclaw/callback.py`、`app.py` | flow test + audit assertions需补充 | Console audit view | Implemented / Partial test |
 | UR-013 密钥安全 | `02-TARGET-ARCHITECTURE.md` §6；ADR-006 | `FileSigningMaterialResolver`、credential refs、Secret Gate | permission/secret-scan test需补充 | env template + external protected file | Implemented / Partial test |
 | UR-014 部署与回滚 | `03-MIGRATION-AND-DELIVERY.md` §7 | deployment scripts/assets | acceptance runner + healthcheck | `deploy/aios-v2/` | Implemented / Pre-prod pending |
 | UR-015 文档与代码同步 | `03-MIGRATION-AND-DELIVERY.md` §9；ADR-005 | `tools/aios_v2_docs_code_check.py` | checker self-test / acceptance runner | CI/Gitea Actions 可调用 | Implemented after checker commit |
 ## 3. API 追踪
 | API | 需求 | 实现 | 测试状态 |
 |---|---|---|---|
 | `GET /api/v2/health` | UR-004, UR-014 | `gateway.py` | healthcheck 定义完成 |
 | `POST /api/v2/tenants` | UR-001, UR-005 | `gateway.py` / `control_plane.py` | 需增加 HTTP 集成测试 |
 | `POST /api/v2/agents` | UR-001, UR-005 | `gateway.py` / `control_plane.py` | Domain flow 已覆盖 |
 | `POST /api/v2/tasks` | UR-006, UR-008 | `gateway.py` / `control_plane.py` | Domain flow 已覆盖 |
 | `POST /api/v2/tasks/{id}/approve` | UR-006, UR-007 | `gateway.py` | Domain flow 已覆盖 |
 | `POST /api/v2/tasks/{id}/dispatch` | UR-001, UR-003 | `app.py` / `orchestrator.py` | OpenClaw/Hermes flow 已覆盖 |
 | `GET /api/v2/runtime` | UR-004, UR-009 | `app.py` | HTTP smoke test待执行 |
 | `GET /api/v2/executions` | UR-009 | `app.py` / `execution_tracking.py` | Evidence persistence test已定义 |
 | `GET /api/v2/gitops` | UR-010 | `app.py` / `gitops_status.py` | Runtime/healthcheck验证待执行 |
 | `POST /api/v2/hermes/results/reconcile` | UR-003, UR-006 | `app.py` / `orchestrator.py` | Hermes flow已覆盖 Domain 路径 |
 ## 4. 运行时契约追踪
 | 契约 | 设计 | 代码 | 测试 |
 |---|---|---|---|
 | OpenClawCommand | `04-MODULE-CONTRACTS.md` §7.1 | `aios_openclaw/contract.py` | OpenClaw full-flow；签名负向测试需补充 |
 | OpenClawResult | §7.2 | `aios_openclaw/contract.py` | OpenClaw full-flow |
 | RuntimeHealth | §7.3 | `aios_openclaw/contract.py`、`health.py` | health policy测试需补充 |
 | HermesJob | §8.1 | `aios_hermes/contract.py` | Hermes full-flow |
 | HermesResult | §8.2 | `aios_hermes/contract.py` | Hermes full-flow |
 | HermesHealth | §8.3 | `aios_hermes/contract.py`、`adapter.py` | healthcheck/worker test待执行 |
 | ExecutionEvidence | §9 | `execution_tracking.py` | evidence persistence test |
 | GitOpsStatus | §10 | `gitops_status.py` | healthcheck detached/clean检查 |
 ## 5. 部署追踪
 | 能力 | 文件 | 验收命令 |
 |---|---|---|
 | systemd | `deploy/aios-v2/systemd/aios-v2.service` | `systemctl status aios-v2` |
 | 环境模板 | `deploy/aios-v2/env/aios-v2.env.example` | `systemd-analyze verify` + 文件权限检查 |
 | Docker | `deploy/aios-v2/docker/Dockerfile` | `docker build` |
 | Compose | `deploy/aios-v2/docker/compose.yaml` | `docker compose config` / `up` |
 | 健康检查 | `deploy/aios-v2/healthcheck/check.sh` | 直接执行 |
 | 回滚 | `deploy/aios-v2/rollback/rollback.sh` | 预发布回滚演练 |
 | 全量验收 | `deploy/aios-v2/acceptance/run-all.sh` | 直接执行 |
 ## 6. 文档覆盖追踪
 | 文档 | 覆盖内容 | 代码/变更触发条件 |
 |---|---|---|
 | `README.md` | 事实源索引、状态语言、实现地图 | 新模块/目录/版本 |
 | `01-USER-REQUIREMENTS.md` | 用户需求、NFR、验收场景 | 用户行为、功能范围、验收变化 |
 | `02-TARGET-ARCHITECTURE.md` | 分层、边界、数据流、安全、租户 | 架构边界、数据流、运行时变化 |
 | `03-MIGRATION-AND-DELIVERY.md` | 迁移、GitOps、Gate、回滚 | 发布流程、迁移阶段、门禁变化 |
 | `04-MODULE-CONTRACTS.md` | 字段、API、状态机、错误语义 | Contract/API/模型变化 |
 | `05-TRACEABILITY-MATRIX.md` | 需求到代码测试部署映射 | 代码、测试、部署路径变化 |
 | `adr/*` | 关键不可逆或跨模块决策 | 新架构决策或旧决策替换 |
 ## 7. 当前未完成验证项
 以下项目代码或测试定义已存在，但尚不能标记为 `Test-verified` 或 `Pre-production verified`：
 1. 精确 Candidate commit 的测试执行结果；
 2. HTTP API 与 Web Console 冒烟测试；
 3. 真实 OpenClaw Runtime 签名、取消、超时和健康联调；
 4. Hermes 长任务心跳、Worker 崩溃和 Lease 恢复演练；
 5. 多实例 SQLite/并发容量边界；
 6. systemd、Docker、Compose、restart 和 rollback 演练；
 7. local/Gitea Candidate OID 一致证明；
 8. 生产身份代理、防伪 Header、CSRF/CORS/Rate Limit 验证；
 9. 文档—代码同步检查器在 CI/Acceptance 中执行通过。
 ## 8. 更新要求
 任何代码、API、部署或测试路径变化，必须同步更新本矩阵。验收脚本会检查本文档引用的关键路径存在，并检查架构/功能代码变化是否同时包含 `docs/aios-v2/` 变化。
@@ -0,0 +1,98 @@
 # AIOS V2 Architecture and Delivery Index
 - Version line: `V2R1C0-candidate`
 - Candidate branch: `integration/aios-v2r1-cto-production-candidate-20260720`
 - Architecture owner: AIOS V2 CTO
 - Status: implementation-complete candidate; runtime and production verification pending
 ## 1. Purpose
 This directory is the single source of truth for AIOS V2 requirements, target architecture, module boundaries, migration strategy, release gates and architectural decisions.
 The design is based on the following non-negotiable principles:
 1. Do not rewrite the existing platform.
 2. Reuse OpenClaw, Hermes and Agent Router capabilities.
 3. Decouple OpenClaw and Hermes from control-plane ownership.
 4. AIOS is the only control plane.
 5. All user-facing and operational capabilities are Web/API accessible.
 6. Every module enforces multi-tenant isolation.
 7. Every change flows through Git branches, commits, tests, review and controlled promotion.
 ## 2. Document map
 | Document | Purpose |
 |---|---|
 | [01-USER-REQUIREMENTS.md](01-USER-REQUIREMENTS.md) | Product goals, user requirements, non-functional requirements and acceptance criteria |
 | [02-TARGET-ARCHITECTURE.md](02-TARGET-ARCHITECTURE.md) | Target logical architecture, control boundaries, data flow, security and tenancy model |
 | [03-MIGRATION-AND-DELIVERY.md](03-MIGRATION-AND-DELIVERY.md) | Incremental migration plan, reuse strategy, GitOps workflow and release gates |
 | [04-MODULE-CONTRACTS.md](04-MODULE-CONTRACTS.md) | Tenant context, task lifecycle, OpenClaw, Hermes, audit and execution-evidence contracts |
 | [05-TRACEABILITY-MATRIX.md](05-TRACEABILITY-MATRIX.md) | Requirement-to-design-to-code-to-test mapping |
 | [adr/README.md](adr/README.md) | Architecture decision records |
 ## 3. Current implementation map
 | Architecture capability | Candidate implementation |
 |---|---|
 | Tenant, agent, task, policy and audit control plane | `active-runtime/app/aios_v2/control_plane.py` |
 | Durable control-plane state | `active-runtime/app/aios_v2/persistence.py` |
 | Web/API gateway | `active-runtime/app/aios_v2/gateway.py` |
 | Unified application entrypoint | `active-runtime/app/aios_v2/app.py` |
 | Runtime dispatch and reconciliation | `active-runtime/app/aios_v2/orchestrator.py` |
 | Agent execution evidence | `active-runtime/app/aios_v2/execution_tracking.py` |
 | GitOps runtime visibility | `active-runtime/app/aios_v2/gitops_status.py` |
 | Web Console | `active-runtime/app/aios_v2/console.py` |
 | OpenClaw execution boundary | `active-runtime/app/aios_openclaw/` |
 | Hermes runtime boundary | `active-runtime/app/aios_hermes/` |
 | Candidate tests | `tests/aios_v2_candidate/` |
 | Deployment, health and rollback | `deploy/aios-v2/` |
 ## 4. Status vocabulary
 The project separates code availability from release readiness:
 - **Designed**: documented boundary and acceptance criteria exist.
 - **Implemented**: code is committed on the candidate branch.
 - **Test-defined**: automated tests exist in Git.
 - **Test-verified**: tests were executed for the exact commit and evidence was recorded.
 - **Pre-production verified**: deployed in an isolated environment with health, restart and rollback evidence.
 - **Production released**: owner-approved merge/tag/deployment completed.
 A branch or file existing does not, by itself, mean a capability is production-ready.
 ## 5. Design summary
 ```text
 Browser / Web Console / Public API
                 |
                 v
          AIOS Control Plane
  Tenant | Agent | Task | Policy | Audit
                 |
                 v
       Execution Orchestrator / Router
             /                 \
            v                   v
   OpenClaw Adapter       Hermes Boundary
            |                   |
            v                   v
   OpenClaw Runtime       Hermes Workers
                 \
                  v
      Result Reconciliation + Evidence
 ```
 The control plane owns authoritative identity, policy, approval, task lifecycle and audit state. OpenClaw and Hermes receive only tenant-scoped, approved execution work and return runtime results.
 ## 6. Release rule
 No candidate may be promoted until:
 - requirement-to-code traceability is complete;
 - multi-tenant negative tests pass;
 - controller workspace is detached and clean;
 - local Git and Gitea branch OIDs match;
 - persistence restart recovery passes;
 - healthcheck and rollback pass in pre-production;
 - no plaintext secret enters Git;
 - owner approval is recorded for merge, tag and production deployment.
@@ -0,0 +1,39 @@
 # ADR-001: AIOS 作为唯一控制面
 - Status: Accepted
 - Date: 2026-07-20
 ## Context
 现有系统包含多个执行和调度能力，若每个 Runtime 保存自己的任务、租户和策略状态，会形成多个事实源。
 ## Decision
 AIOS Control Plane 是唯一权威控制面，负责：
 - Tenant
 - Agent
 - Task
 - Policy
 - Approval
 - Audit
 - Execution Evidence
 OpenClaw、Hermes、Agent Router 只能执行或路由，不拥有第二套控制状态。
 ## Consequences
 Positive:
 - 状态唯一；
 - 审计完整；
 - Web/API 统一。
 Negative:
 - Runtime 接入需要 Adapter。
 - 迁移需要阶段性边界收敛。
 ## Compliance
 任何新 Runtime 必须通过 AIOS Orchestrator 接入。
@@ -0,0 +1,46 @@
 # ADR-002: OpenClaw/Hermes 与控制面解耦
 - Status: Accepted
 - Date: 2026-07-20
 ## Context
 OpenClaw 和 Hermes 已具备执行能力，但历史上可能承担部分控制职责。
 ## Decision
 引入边界层：
 ```text
 AIOS Control Plane
        |
        v
 Adapter / Boundary
        |
        v
 Runtime
 ```
 OpenClaw：
 - 执行签名命令；
 - 返回结果；
 - 提供健康状态。
 Hermes：
 - 队列；
 - Worker；
 - Lease；
 - Heartbeat；
 - 执行结果。
 ## Consequences
 - Runtime 可独立演进。
 - 控制逻辑不会分裂。
 - 需要维护稳定 Contract。
 ## Migration
 先接 Adapter，再迁移旧调用入口，最后关闭控制旁路。
@@ -0,0 +1,44 @@
 # ADR-003: 所有变更必须通过 GitOps
 - Status: Accepted
 - Date: 2026-07-20
 ## Context
 直接在生产目录、长期 worktree 或主分支上修改代码，会造成不可追踪、不可回滚和分支冲突。
 ## Decision
 所有代码、配置、文档、测试、部署和回滚资产必须通过以下流程：
 ```text
 non-main branch -> commit -> local sync -> Gitea sync -> test -> review -> controlled promotion
 ```
 Controller worktree 必须长期保持 detached 且 clean。每个写任务使用独立工作区或受控写入流程；相同分支写入必须串行。
 ## Required evidence
 每个任务必须记录：
 - branch
 - commit
 - changed files
 - test command and result
 - blocker and resolution
 - local/Gitea sync status
 - rollback note
 ## Prohibited
 - 直接 push `main`
 - 未审批创建生产 Tag
 - 在生产目录手工修改后不回写 Git
 - fetch 到已被 worktree 检出的本地 branch ref
 - 在 Git 中提交明文密钥
 ## Consequences
 优点：可审计、可回滚、可并行治理。
 成本：紧急修复也必须形成 patch branch 和证据。
@@ -0,0 +1,43 @@
 # ADR-005: 文档与代码同步门禁
 - Status: Accepted
 - Date: 2026-07-20
 ## Context
 架构、功能和部署变化如果不同步文档，会导致版本无法审计，后续 Agent 无法判断真实设计。
 ## Decision
 所有影响以下内容的变更必须同步 `docs/aios-v2/`：
 - 用户行为；
 - API；
 - 数据模型；
 - 状态机；
 - Runtime Boundary；
 - 安全策略；
 - 部署流程；
 - 测试门禁。
 ## Gate
 验收必须检查：
 1. 文档引用代码路径存在；
 2. API 文档有实现映射；
 3. Contract 字段有模型映射；
 4. Traceability Matrix 已更新；
 5. ADR 与架构变化一致。
 ## Consequences
 优点：
 - 架构长期可维护；
 - Agent 执行有明确上下文；
 - Bug 修复有历史依据。
 成本：
 - 小功能也需要更新对应文档。
@@ -0,0 +1,39 @@
 # AIOS V2 Architecture Decision Records
 本目录记录影响多个模块、长期演进、兼容性、安全或交付流程的关键架构决策。
 ## ADR 列表
 | ADR | 决策 |
 |---|---|
 | [ADR-001](ADR-001-single-control-plane.md) | AIOS 作为唯一控制面 |
 | [ADR-002](ADR-002-openclaw-hermes-decoupling.md) | OpenClaw/Hermes 与控制面解耦 |
 | [ADR-003](ADR-003-gitops-only-change-flow.md) | 所有变更必须通过 GitOps |
 | [ADR-004](ADR-004-multi-tenant-isolation.md) | 多租户隔离默认拒绝 |
 | [ADR-005](ADR-005-document-code-sync-gate.md) | 文档与代码同步门禁 |
 | [ADR-006](ADR-006-candidate-persistence-strategy.md) | 候选持久化采用可替换 SQLite Snapshot |
 ## ADR 状态
 - `Proposed`：提出但未采纳。
 - `Accepted`：当前有效决策。
 - `Superseded`：被新 ADR 替代。
 - `Deprecated`：保留历史但禁止新实现采用。
 ## 何时必须新增或更新 ADR
 以下变化必须新增或更新 ADR：
 1. 控制面与 Runtime 权限边界变化；
 2. 多租户隔离模型变化；
 3. API、数据模型或状态机出现不兼容变化；
 4. 持久化、消息、密钥、审计或部署方案发生跨模块变化；
 5. GitOps、发布、回滚或文档门禁变化；
 6. 替换 OpenClaw、Hermes、Agent Router 或核心基础设施。
 ADR 变化必须同步更新：
 - `02-TARGET-ARCHITECTURE.md`
 - `04-MODULE-CONTRACTS.md`
 - `05-TRACEABILITY-MATRIX.md`
 - 对应代码、测试和部署资产
@@ -1,65 +0,0 @@
 # A0 callback transport and pending-decision recovery
 This runbook separates code acceptance from live callback acceptance. Passing
 unit tests does not prove that a cloud Workspace Agent can reach the installed
 tool endpoint.
 ## Transport configuration
 Choose exactly one mode in the owner-managed remote-tool configuration:
 - `remote_https_action`: `endpoint` must be the real public HTTPS Action URL.
 - `secure_mcp_tunnel`: `endpoint` must exactly match an entry in
  `approved_mcp_tunnel_endpoints`.
 The configuration stores only `access_secret_ref` and `callback_secret_ref`.
 It must not contain resolved credentials. Startup rejects HTTP, loopback,
 unspecified/private literal addresses, URL credentials, query/fragment data,
 unapproved tunnel endpoints, and known placeholder markers.
 Generate the cloud-facing manifest only after owner configuration exists:
 ```bash
 PYTHONPATH=/opt/aios-orchestrator-gateway/current \
 python3 /opt/aios-orchestrator-gateway/current/scripts/generate_orchestrator_tool_manifest.py \
  --transport-config /etc/aios-orchestrator-gateway/remote-tool.json \
  --output /var/lib/aios-orchestrator-gateway/orchestrator-tool-manifest.json
 ```
 The generated Action manifest exposes only
 `submit_orchestration_decision`. The MCP manifest lists the same single tool
 with write-side-effect and idempotency annotations.
 ## Recovery
 Plan mode is read-only:
 ```bash
 PYTHONPATH=/opt/aios-orchestrator-gateway/current \
 python3 /opt/aios-orchestrator-gateway/current/scripts/recover_pending_decisions.py \
  --config /etc/aios-orchestrator-gateway/config.json --plan
 ```
 Application handles at most one leased record and never resends a model prompt
 or replays a decision:
 ```bash
 PYTHONPATH=/opt/aios-orchestrator-gateway/current \
 python3 /opt/aios-orchestrator-gateway/current/scripts/recover_pending_decisions.py \
  --config /etc/aios-orchestrator-gateway/config.json --apply-one
 ```
 ## Required live acceptance
 Do not mark the live Gate PASS until metadata-only evidence proves, in order:
 1. Workspace trigger returned HTTP 202.
 2. The remote callback tool received one authenticated request.
 3. The bridge returned HTTP 202 from the loopback 18920 decision endpoint.
 4. The event reached `DECIDED`, `callback_state=APPLIED`, and
   `decision_applied=true`.
 5. Recovery planning and one apply cycle did not create a second outbox side
   effect for the same event.
 Never include Authorization, token values, resolved SecretRefs, prompts, or raw
 model output in logs or RESULT files.
@@ -0,0 +1,47 @@
 # TASK-AIOS-CONTROL-PLANE-GATEWAY-V1
 ## Owner
 Codex
 ## Goal
 建立 AIOS V2 唯一控制面 Gateway v1。
 ## Architecture Boundary
 AIOS Control Plane:
 - Tenant Registry
 - Agent Registry
 - Task Lifecycle
 - Policy Gate
 - Approval
 - Audit Event
 不允许：
 - OpenClaw 直接管理任务生命周期
 - Hermes 直接暴露控制 API
 - Web UI 绕过 AIOS API
 ## Deliverables
 1. Control Plane API skeleton
 2. Tenant context middleware
 3. Task state machine
 4. Agent registration contract
 5. Audit event interface
 ## Acceptance
 - 所有 API 携带 tenant_id
 - 未授权跨租户访问拒绝
 - Task create/update/cancel 可审计
 - GitOps commit 可追踪
 - pytest + integration test 通过
 ## Migration
 复用：
 - Agent Router
 - OpenClaw adapter
 - Hermes runtime
 不重写执行层。
@@ -1,89 +0,0 @@
 openapi: 3.0.3
 info:
  title: AIOS Orchestrator Gateway metadata API
  version: 0.2.0
 servers:
  - url: http://127.0.0.1:18920
 paths:
  /v1/orchestrator/events:
    post:
      security:
        - bearerAuth: []
      requestBody:
        required: true
        content:
          application/json:
            schema:
              type: object
      responses:
        '202':
          description: Event accepted
        '409':
          description: Idempotency or state conflict
        '422':
          description: Invalid event
  /v1/orchestrator/events/{event_id}:
    get:
      security:
        - bearerAuth: []
      parameters:
        - in: path
          name: event_id
          required: true
          schema:
            type: string
            pattern: '^[A-Za-z0-9][A-Za-z0-9._:-]{0,127}$'
      responses:
        '200':
          description: Metadata-only event status
          content:
            application/json:
              schema:
                type: object
                additionalProperties: false
                required: [event_id, status, trigger_http_status, trigger_error_category, trigger_channel, remote_id, callback_state, callback_deadline, callback_attempts, decision_applied, state_version_after]
                properties:
                  event_id:
                    type: string
                  status:
                    enum: [RECEIVED, AWAITING_DECISION, DECIDED, TRIGGER_FAILED, DEAD_LETTER]
                  trigger_channel:
                    type: string
                    nullable: true
                    enum: [workspace_agent, responses_api]
                  trigger_http_status:
                    type: integer
                    nullable: true
                    minimum: 100
                    maximum: 599
                  trigger_error_category:
                    type: string
                    nullable: true
                    enum: [UNAUTHORIZED, FORBIDDEN, NOT_FOUND, NOT_RUNNABLE, NETWORK_ERROR]
                  remote_id:
                    type: string
                    nullable: true
                  callback_state:
                    enum: [PENDING, SENT, AWAITING_CALLBACK, CALLBACK_RECEIVED, APPLIED, STALE_SUPERSEDED, DEAD_LETTER]
                  callback_deadline:
                    type: integer
                    nullable: true
                    minimum: 0
                  callback_attempts:
                    type: integer
                    minimum: 0
                  decision_applied:
                    type: boolean
                  state_version_after:
                    type: integer
                    nullable: true
                    minimum: 0
        '401':
          description: Missing or invalid bearer
        '404':
          description: Event not found
 components:
  securitySchemes:
    bearerAuth:
      type: http
      scheme: bearer
@@ -0,0 +1,9 @@
 {
  "type": "branch",
  "branch": "integration/aios-v2r1-cto-production-candidate-20260720",
  "base": "task/codex/AIOS-V2-CONTROL-PLANE-GATEWAY-V1-20260720",
  "created_by": "chatgpt-gitops-bot",
  "created_at": "2026-07-20T01:47:01Z",
  "direct_main_push": false,
  "owner_approval_required": true
 }
@@ -0,0 +1,9 @@
 {
  "type": "branch",
  "branch": "task/codex/AIOS-V2-CONTROL-PLANE-GATEWAY-V1-20260720",
  "base": "main",
  "created_by": "chatgpt-gitops-bot",
  "created_at": "2026-07-19T17:06:04Z",
  "direct_main_push": false,
  "owner_approval_required": true
 }
@@ -0,0 +1,9 @@
 {
  "type": "branch",
  "branch": "task/openclaw/AIOS-V2R1-PRODUCTION-DEPLOY-20260720",
  "base": "integration/aios-v2r1-cto-production-candidate-20260720",
  "created_by": "chatgpt-gitops-bot",
  "created_at": "2026-07-20T03:49:09Z",
  "direct_main_push": false,
  "owner_approval_required": true
 }
@@ -1,61 +0,0 @@
 {
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "$id": "https://aios.local/schemas/automation-quorum-controller.schema.json",
  "title": "AIOS GPT plus one-executor quorum controller configuration",
  "type": "object",
  "additionalProperties": false,
  "required": [
    "project_key",
    "contract_commit",
    "database_path",
    "file_inbox_path",
    "health_receipts",
    "blocker_retry_threshold",
    "max_health_age_seconds",
    "lease_ttl_seconds",
    "gateway_event_url",
    "gateway_token_file"
  ],
  "properties": {
    "$schema": {"type": "string"},
    "project_key": {"type": "string", "pattern": "^[A-Za-z0-9][A-Za-z0-9._:-]{0,127}$"},
    "contract_commit": {"type": "string", "pattern": "^[0-9a-f]{40}$"},
    "database_path": {"type": "string", "pattern": "^/"},
    "file_inbox_path": {"type": "string", "pattern": "^/"},
    "default_workload": {
      "enum": [
        "large_repetitive_infrastructure_monitoring",
        "complex_code_security_architecture"
      ]
    },
    "health_receipts": {
      "type": "object",
      "additionalProperties": false,
      "required": ["central_gpt", "openclaw", "codex"],
      "properties": {
        "central_gpt": {"type": "string", "pattern": "^/"},
        "openclaw": {"type": "string", "pattern": "^/"},
        "codex": {"type": "string", "pattern": "^/"}
      }
    },
    "git_ref_polling": {
      "type": "object",
      "additionalProperties": false,
      "required": ["repository", "refs"],
      "properties": {
        "repository": {"type": "string", "pattern": "^/"},
        "refs": {
          "type": "array",
          "minItems": 1,
          "uniqueItems": true,
          "items": {"type": "string", "pattern": "^refs/"}
        }
      }
    },
    "blocker_retry_threshold": {"type": "integer", "minimum": 1, "maximum": 20},
    "max_health_age_seconds": {"type": "integer", "minimum": 5, "maximum": 300},
    "lease_ttl_seconds": {"type": "integer", "minimum": 15, "maximum": 3600},
    "gateway_event_url": {"type": "string", "const": "http://127.0.0.1:18920/v1/orchestrator/events"},
    "gateway_token_file": {"type": "string", "pattern": "^/"}
  }
 }
@@ -1,22 +0,0 @@
 {
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "title": "AIOS Callback Transport",
  "type": "object",
  "additionalProperties": false,
  "required": ["mode", "endpoint", "approved_mcp_tunnel_endpoints"],
  "properties": {
    "mode": {
      "enum": ["remote_https_action", "secure_mcp_tunnel"]
    },
    "endpoint": {
      "type": "string",
      "format": "uri",
      "pattern": "^https://"
    },
    "approved_mcp_tunnel_endpoints": {
      "type": "array",
      "uniqueItems": true,
      "items": {"type": "string", "format": "uri", "pattern": "^https://"}
    }
  }
 }
@@ -1,25 +0,0 @@
 {
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "$id": "https://aios.local/schemas/codex-command.schema.json",
  "title": "AIOS Codex command",
  "type": "object",
  "additionalProperties": false,
  "required": ["schema_version", "command_id", "project_key", "work_unit_id", "contract_commit", "state_version", "execution"],
  "properties": {
    "schema_version": {"const": 1},
    "command_id": {"type": "string", "pattern": "^[A-Za-z0-9][A-Za-z0-9._:-]{0,127}$"},
    "project_key": {"type": "string", "pattern": "^[A-Za-z0-9][A-Za-z0-9._:-]{0,127}$"},
    "work_unit_id": {"type": "string", "pattern": "^[A-Za-z0-9][A-Za-z0-9._:-]{0,127}$"},
    "contract_commit": {"type": "string", "pattern": "^[0-9a-f]{40}$"},
    "state_version": {"type": "integer", "minimum": 0},
    "execution": {
      "type": "object",
      "additionalProperties": false,
      "required": ["kind", "input"],
      "properties": {
        "kind": {"enum": ["acceptance_probe", "codex"]},
        "input": {"type": "string", "maxLength": 65536}
      }
    }
  }
 }
@@ -1,145 +0,0 @@
 {
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "title": "AIOS Orchestrator Decision",
  "type": "object",
  "additionalProperties": false,
  "required": [
    "schema_version",
    "project_key",
    "contract_commit",
    "event_id",
    "expected_state_version",
    "verdict",
    "state_updates",
    "next_work_units",
    "requires_human_input",
    "stop_reason"
  ],
  "properties": {
    "schema_version": {
      "type": "integer",
      "const": 1
    },
    "project_key": {
      "type": "string"
    },
    "contract_commit": {
      "type": "string",
      "pattern": "^[0-9a-f]{40}$"
    },
    "event_id": {
      "type": "string"
    },
    "expected_state_version": {
      "type": "integer",
      "minimum": 0
    },
    "verdict": {
      "type": "string",
      "enum": [
        "BLOCKED",
        "COMPLETE",
        "DISPATCH",
        "NOOP"
      ]
    },
    "state_updates": {
      "type": "array",
      "items": {
        "type": "object",
        "additionalProperties": false,
        "required": [
          "key",
          "value"
        ],
        "properties": {
          "key": {
            "type": "string"
          },
          "value": {
            "type": "string"
          }
        }
      }
    },
    "next_work_units": {
      "type": "array",
      "maxItems": 32,
      "items": {
        "type": "object",
        "additionalProperties": false,
        "required": [
          "id",
          "track_id",
          "gate_id",
          "primary_agent",
          "fallback_agent",
          "clean_base_commit",
          "allowed_paths",
          "required_checks",
          "forbidden_operations",
          "instruction"
        ],
        "properties": {
          "id": {
            "type": "string"
          },
          "track_id": {
            "type": "string"
          },
          "gate_id": {
            "type": "string"
          },
          "primary_agent": {
            "type": "string",
            "enum": [
              "openclaw",
              "codex"
            ]
          },
          "fallback_agent": {
            "type": "string",
            "enum": [
              "openclaw",
              "codex"
            ]
          },
          "clean_base_commit": {
            "type": "string",
            "pattern": "^[0-9a-f]{40}$"
          },
          "allowed_paths": {
            "type": "array",
            "items": {
              "type": "string"
            }
          },
          "required_checks": {
            "type": "array",
            "items": {
              "type": "string"
            }
          },
          "forbidden_operations": {
            "type": "array",
            "items": {
              "type": "string"
            }
          },
          "instruction": {
            "type": "string"
          }
        }
      }
    },
    "requires_human_input": {
      "type": "boolean"
    },
    "stop_reason": {
      "type": [
        "string",
        "null"
      ]
    }
  }
 }
@@ -1,102 +0,0 @@
 {
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "title": "AIOS Orchestrator Event",
  "type": "object",
  "additionalProperties": false,
  "required": [
    "schema_version",
    "project_key",
    "contract_commit",
    "event_id",
    "source_agent",
    "event_type",
    "work_unit_id",
    "status",
    "state_version",
    "implementation_commit",
    "result_commit",
    "blocker",
    "evidence_refs",
    "payload"
  ],
  "properties": {
    "schema_version": {
      "const": 1
    },
    "project_key": {
      "type": "string",
      "minLength": 1,
      "maxLength": 128
    },
    "contract_commit": {
      "type": "string",
      "pattern": "^[0-9a-f]{40}$"
    },
    "event_id": {
      "type": "string",
      "minLength": 1,
      "maxLength": 128
    },
    "source_agent": {
      "enum": [
        "openclaw",
        "codex",
        "gitea",
        "system",
        "workspace-agent"
      ]
    },
    "event_type": {
      "type": "string"
    },
    "work_unit_id": {
      "type": "string"
    },
    "status": {
      "enum": [
        "READY",
        "RUNNING",
        "PASS",
        "BLOCKED",
        "FAIL",
        "TIMEOUT",
        "NOT_EXECUTED"
      ]
    },
    "state_version": {
      "type": "integer",
      "minimum": 0
    },
    "implementation_commit": {
      "type": [
        "string",
        "null"
      ],
      "pattern": "^[0-9a-f]{40}$"
    },
    "result_commit": {
      "type": [
        "string",
        "null"
      ],
      "pattern": "^[0-9a-f]{40}$"
    },
    "blocker": {
      "type": [
        "string",
        "null"
      ],
      "maxLength": 512
    },
    "evidence_refs": {
      "type": "array",
      "items": {
        "type": "string"
      },
      "maxItems": 64
    },
    "payload": {
      "type": "object"
    }
  }
 }
@@ -1,63 +0,0 @@
 #!/usr/bin/env python3
 """Seed exactly one orchestration event when the project has no durable state."""
 from __future__ import annotations
 import argparse
 from pathlib import Path
 import sys
 REPO_ROOT = Path(__file__).resolve().parents[1]
 sys.path.insert(0, str(REPO_ROOT))
 from services.aios_orchestrator_gateway.config import Config
 from services.aios_orchestrator_gateway.models import Event
 from services.aios_orchestrator_gateway.securefs import atomic_write_bytes_shared
 from services.aios_orchestrator_gateway.store import Store
 def main() -> int:
    parser = argparse.ArgumentParser()
    parser.add_argument("--config", type=Path, required=True)
    args = parser.parse_args()
    config = Config.from_file(args.config)
    store = Store(config.database_path)
    if store.project_state(config.project_key) is not None:
        print("ORCHESTRATION_BOOTSTRAP=NOOP reason=STATE_EXISTS")
        return 0
    event = Event.from_dict(
        {
            "schema_version": 1,
            "project_key": config.project_key,
            "contract_commit": config.contract_commit,
            "event_id": "bootstrap-" + config.contract_commit[:24],
            "source_agent": "system",
            "event_type": "orchestration_bootstrap",
            "work_unit_id": "ORCHESTRATION-BOOTSTRAP",
            "status": "READY",
            "state_version": 0,
            "implementation_commit": None,
            "result_commit": None,
            "blocker": None,
            "evidence_refs": [],
            "payload": {
                "reason": "initial_gateway_install",
                "polling_enabled": True,
                "openclaw_sync_required": True,
            },
        }
    )
    path = config.event_inbox_dir / "pending" / f"{event.event_id}.json"
    atomic_write_bytes_shared(
        path,
        (event.canonical_json() + "\n").encode(),
        mode=0o640,
        group_writable_directory=True,
    )
    print(f"ORCHESTRATION_BOOTSTRAP=PASS event_id={event.event_id}")
    return 0
 if __name__ == "__main__":
    raise SystemExit(main())
@@ -1,56 +0,0 @@
 #!/usr/bin/env python3
 from __future__ import annotations
 import argparse
 import json
 import os
 from pathlib import Path
 import tempfile
 from services.aios_orchestrator_gateway.callback_transport import CallbackTransport
 def atomic_write(path: Path, value: dict) -> None:
    path.parent.mkdir(parents=True, exist_ok=True)
    fd, temporary = tempfile.mkstemp(prefix=f".{path.name}.", dir=path.parent)
    try:
        with os.fdopen(fd, "w", encoding="utf-8") as handle:
            json.dump(value, handle, indent=2, sort_keys=True)
            handle.write("\n")
            handle.flush()
            os.fsync(handle.fileno())
        os.chmod(temporary, 0o600)
        os.replace(temporary, path)
        directory = os.open(path.parent, os.O_RDONLY | os.O_DIRECTORY)
        try:
            os.fsync(directory)
        finally:
            os.close(directory)
    finally:
        try:
            os.unlink(temporary)
        except FileNotFoundError:
            pass
 def main() -> int:
    parser = argparse.ArgumentParser()
    parser.add_argument("--transport-config", type=Path, required=True)
    parser.add_argument("--output", type=Path, required=True)
    args = parser.parse_args()
    value = json.loads(args.transport_config.read_text(encoding="utf-8"))
    if "callback_transport" in value:
        value = value["callback_transport"]
    transport = CallbackTransport.from_dict(value)
    manifest = (
        transport.action_openapi()
        if transport.mode == "remote_https_action"
        else {"endpoint": transport.endpoint, "tools": [transport.mcp_tool()]}
    )
    atomic_write(args.output, manifest)
    print(f"TOOL_MANIFEST=PASS mode={transport.mode}")
    return 0
 if __name__ == "__main__":
    raise SystemExit(main())
@@ -1,64 +0,0 @@
 #!/usr/bin/env bash
 set -euo pipefail
 [[ ${EUID} -eq 0 ]] || { echo "root required" >&2; exit 1; }
 ROOT=$(cd -- "$(dirname -- "${BASH_SOURCE[0]}")/.." && pwd -P)
 CONFIG=${1:-/etc/aios-codex-watcher/config.json}
 [[ "$CONFIG" = /* ]] || { echo "config path must be absolute" >&2; exit 1; }
 [[ -f "$CONFIG" && ! -L "$CONFIG" ]] || { echo "pre-provisioned regular config required" >&2; exit 1; }
 for path in \
  services/aios_codex_watcher/__init__.py \
  services/aios_codex_watcher/__main__.py \
  services/aios_codex_watcher/watcher.py \
  systemd/aios-codex-watcher.service \
  systemd/aios-codex-watcher.timer \
  systemd/aios-codex-watcher.sysusers \
  systemd/aios-codex-watcher.tmpfiles; do
  [[ -f "$ROOT/$path" && ! -L "$ROOT/$path" ]] || { echo "missing or unsafe: $path" >&2; exit 1; }
 done
 digest=$(sha256sum \
  "$ROOT/services/aios_codex_watcher/__init__.py" \
  "$ROOT/services/aios_codex_watcher/__main__.py" \
  "$ROOT/services/aios_codex_watcher/watcher.py" | sha256sum | awk '{print $1}')
 release="/opt/aios-codex-watcher/releases/$digest"
 install -d -o root -g root -m 0755 "$release/services/aios_codex_watcher" "$release/bin"
 install -o root -g root -m 0644 "$ROOT/services/aios_codex_watcher/"*.py "$release/services/aios_codex_watcher/"
 launcher="$release/bin/aios-codex-watcher"
 printf '%s\n' '#!/usr/bin/env bash' 'set -euo pipefail' \
  "export PYTHONPATH='$release'" \
  "exec /usr/bin/python3 -m services.aios_codex_watcher \"\$@\"" >"$launcher"
 chown root:root "$launcher"
 chmod 0755 "$launcher"
 install -d -o root -g root -m 0755 /opt/aios-codex-watcher /etc/aios-codex-watcher
 ln -sfn "releases/$digest" /opt/aios-codex-watcher/current.new
 mv -Tf /opt/aios-codex-watcher/current.new /opt/aios-codex-watcher/current
 install -o root -g root -m 0644 "$ROOT/systemd/aios-codex-watcher.service" /etc/systemd/system/
 install -o root -g root -m 0644 "$ROOT/systemd/aios-codex-watcher.timer" /etc/systemd/system/
 install -o root -g root -m 0644 "$ROOT/systemd/aios-codex-watcher.sysusers" /usr/lib/sysusers.d/
 install -o root -g root -m 0644 "$ROOT/systemd/aios-codex-watcher.tmpfiles" /usr/lib/tmpfiles.d/
 systemd-sysusers /usr/lib/sysusers.d/aios-codex-watcher.sysusers
 systemd-tmpfiles --create /usr/lib/tmpfiles.d/aios-codex-watcher.tmpfiles
 commands=/data/aios-gitops/workspace/agents/commands
 for directory in "$commands/active" "$commands/completed"; do
  [[ -d "$directory" && ! -L "$directory" ]] || { echo "missing or unsafe command directory: $directory" >&2; exit 1; }
  [[ $(stat -c '%U:%G' "$directory") == aios-orchestrator:aios-orchestrator ]] || { echo "unexpected command directory owner: $directory" >&2; exit 1; }
  mode=$(stat -c '%a' "$directory")
  (( (8#$mode & 2) == 0 )) || { echo "world-writable command directory: $directory" >&2; exit 1; }
 done
 if [[ -e "$commands/dead-letter" || -L "$commands/dead-letter" ]]; then
  [[ -d "$commands/dead-letter" && ! -L "$commands/dead-letter" ]] || { echo "unsafe dead-letter directory" >&2; exit 1; }
 else
  install -d -o aios-orchestrator -g aios-orchestrator -m 0750 "$commands/dead-letter"
 fi
 [[ $(stat -c '%U:%G' "$commands/dead-letter") == aios-orchestrator:aios-orchestrator ]] || { echo "unexpected dead-letter owner" >&2; exit 1; }
 mode=$(stat -c '%a' "$commands/dead-letter")
 (( (8#$mode & 2) == 0 )) || { echo "world-writable dead-letter directory" >&2; exit 1; }
 printf '{"release_sha256":"%s","installed_at":"%s"}\n' "$digest" "$(date -u +%FT%TZ)" >"$release/receipt.json"
 chmod 0644 "$release/receipt.json"
 systemctl daemon-reload
 systemctl enable --now aios-codex-watcher.timer
 printf 'installed_sha256=%s\n' "$digest"
@@ -1,334 +0,0 @@
 #!/usr/bin/env bash
 set -Eeuo pipefail
 umask 0077
 SOURCE_ROOT="${SOURCE_ROOT:-$(cd "$(dirname "${BASH_SOURCE[0]}")/.." && pwd)}"
 INSTALL_ROOT="${INSTALL_ROOT:-/opt/aios-orchestrator-gateway}"
 CONFIG_ROOT="${CONFIG_ROOT:-/etc/aios-orchestrator-gateway}"
 STATE_ROOT="${STATE_ROOT:-/var/lib/aios-orchestrator-gateway}"
 UNIT_ROOT="${UNIT_ROOT:-/etc/systemd/system}"
 SYSUSERS_ROOT="${SYSUSERS_ROOT:-/etc/sysusers.d}"
 TMPFILES_ROOT="${TMPFILES_ROOT:-/etc/tmpfiles.d}"
 DRY_RUN=0
 if [[ "${1:-}" == "--dry-run" ]]; then
  DRY_RUN=1
 fi
 fail() {
  printf 'AIOS_ORCHESTRATOR_INSTALL=BLOCKED reason=%s\n' "$1" >&2
  exit 2
 }
 [[ "$(id -u)" -eq 0 ]] || fail OWNER_INSTALL_REQUIRED
 for cmd in git python3 systemd-analyze systemctl systemd-sysusers systemd-tmpfiles sha256sum find sort xargs; do
  command -v "$cmd" >/dev/null 2>&1 || fail "MISSING_COMMAND_${cmd}"
 done
 if [[ -f "$SOURCE_ROOT/SHA256SUMS" ]]; then
  (cd "$SOURCE_ROOT" && sha256sum --check SHA256SUMS)
 fi
 for path in \
  "$SOURCE_ROOT/services/aios_orchestrator_gateway" \
  "$SOURCE_ROOT/systemd/aios-orchestrator-gateway.service" \
  "$SOURCE_ROOT/systemd/aios-orchestrator-gateway-poll.service" \
  "$SOURCE_ROOT/systemd/aios-orchestrator-gateway-poll.timer" \
  "$SOURCE_ROOT/systemd/aios-orchestrator-gateway.sysusers" \
  "$SOURCE_ROOT/systemd/aios-orchestrator-gateway.tmpfiles" \
  "$SOURCE_ROOT/systemd/aios-orchestrator-remote-tool.service" \
  "$SOURCE_ROOT/config/aios-orchestrator-gateway.example.json"; do
  [[ -e "$path" ]] || fail "MISSING_SOURCE_${path}"
 done
 PYCACHE="/tmp/aios-orchestrator-install-pycache-$$"
 rm -rf "$PYCACHE"
 PYTHONPYCACHEPREFIX="$PYCACHE" python3 -m compileall -q "$SOURCE_ROOT/services" "$SOURCE_ROOT/scripts"
 PYTHONDONTWRITEBYTECODE=1 PYTHONPATH="$SOURCE_ROOT" \
  python3 -m unittest -v tests.test_aios_orchestrator_gateway
 rm -rf "$PYCACHE"
 for script in "$SOURCE_ROOT"/scripts/*.sh; do bash -n "$script"; done
 python3 -m json.tool "$SOURCE_ROOT/config/aios-orchestrator-gateway.example.json" >/dev/null
 systemd-analyze verify \
  "$SOURCE_ROOT/systemd/aios-orchestrator-gateway.service" \
  "$SOURCE_ROOT/systemd/aios-orchestrator-gateway-poll.service" \
  "$SOURCE_ROOT/systemd/aios-orchestrator-gateway-poll.timer"
 SOURCE_FILES=(
  README-AIOS-ORCHESTRATOR-GATEWAY.md
  docs/runbooks/A0-CALLBACK-TRANSPORT-AND-PENDING-RECOVERY.md
  config/aios-orchestrator-gateway.example.json
  schemas/decision.schema.json
  schemas/event.schema.json
  scripts/bootstrap_orchestration_loop.py
  scripts/install_aios_orchestrator_gateway.sh
  scripts/report_orchestrator_event.py
  scripts/rollback_aios_orchestrator_gateway.sh
  scripts/verify_aios_orchestrator_gateway.sh
  systemd/aios-orchestrator-gateway-poll.service
  systemd/aios-orchestrator-gateway-poll.timer
  systemd/aios-orchestrator-gateway.service
  systemd/aios-orchestrator-gateway.sysusers
  systemd/aios-orchestrator-gateway.tmpfiles
  systemd/aios-orchestrator-remote-tool.service
  schemas/callback-transport.schema.json
  scripts/generate_orchestrator_tool_manifest.py
  scripts/recover_pending_decisions.py
  tests/test_aios_orchestrator_gateway.py
  tests/test_aios_orchestrator_gateway_rollback.sh
  tests/test_aios_callback_recovery.py
  workspace-agent/CENTRAL_AGENT_INSTRUCTIONS.md
  workspace-agent/orchestrator-decision-action.openapi.yaml
  openapi/aios-orchestrator-gateway.openapi.yaml
 )
 while IFS= read -r path; do SOURCE_FILES+=("$path"); done < <(
  cd "$SOURCE_ROOT"
  find services/aios_orchestrator_gateway -type f ! -path '*/__pycache__/*' ! -name '*.pyc' -print | sort
 )
 for path in "${SOURCE_FILES[@]}"; do
  [[ -f "$SOURCE_ROOT/$path" && ! -L "$SOURCE_ROOT/$path" ]] || fail "INVALID_SOURCE_FILE_${path}"
 done
 SOURCE_DIGEST="$(
  cd "$SOURCE_ROOT"
  printf '%s\0' "${SOURCE_FILES[@]}" | sort -z | xargs -0 sha256sum \
    | sha256sum \
    | awk '{print $1}'
 )"
 [[ "$SOURCE_DIGEST" =~ ^[0-9a-f]{64}$ ]] || fail SOURCE_DIGEST_INVALID
 RELEASE_DIR="$INSTALL_ROOT/releases/$SOURCE_DIGEST"
 if [[ "$DRY_RUN" -eq 1 ]]; then
  printf 'AIOS_ORCHESTRATOR_INSTALL_DRY_RUN=PASS digest=%s release=%s\n' \
    "$SOURCE_DIGEST" "$RELEASE_DIR"
  exit 0
 fi
 systemd-sysusers "$SOURCE_ROOT/systemd/aios-orchestrator-gateway.sysusers"
 install -d -m 0755 "$INSTALL_ROOT/releases"
 install -d -o root -g aios-orchestrator -m 0750 "$CONFIG_ROOT"
 install -d -m 0700 "$STATE_ROOT"
 systemd-tmpfiles --create "$SOURCE_ROOT/systemd/aios-orchestrator-gateway.tmpfiles"
 if [[ ! -d "$RELEASE_DIR" ]]; then
  TMP_RELEASE="$INSTALL_ROOT/releases/.tmp-$SOURCE_DIGEST-$$"
  install -d -m 0755 "$TMP_RELEASE"
  for path in "${SOURCE_FILES[@]}"; do
    install -d -m 0755 "$TMP_RELEASE/$(dirname "$path")"
    cp -a "$SOURCE_ROOT/$path" "$TMP_RELEASE/$path"
  done
  printf '%s\n' "$SOURCE_DIGEST" > "$TMP_RELEASE/SOURCE_DIGEST"
  chmod -R go-w "$TMP_RELEASE"
  mv "$TMP_RELEASE" "$RELEASE_DIR"
 fi
 PREVIOUS_TARGET=""
 if [[ -L "$INSTALL_ROOT/current" ]]; then
  PREVIOUS_TARGET="$(readlink -f "$INSTALL_ROOT/current")"
 fi
 install -m 0644 "$SOURCE_ROOT/systemd/aios-orchestrator-gateway.service" \
  "$UNIT_ROOT/aios-orchestrator-gateway.service"
 install -m 0644 "$SOURCE_ROOT/systemd/aios-orchestrator-gateway-poll.service" \
  "$UNIT_ROOT/aios-orchestrator-gateway-poll.service"
 install -m 0644 "$SOURCE_ROOT/systemd/aios-orchestrator-gateway-poll.timer" \
  "$UNIT_ROOT/aios-orchestrator-gateway-poll.timer"
 install -m 0644 "$SOURCE_ROOT/systemd/aios-orchestrator-gateway.sysusers" \
  "$SYSUSERS_ROOT/aios-orchestrator-gateway.conf"
 install -m 0644 "$SOURCE_ROOT/systemd/aios-orchestrator-gateway.tmpfiles" \
  "$TMPFILES_ROOT/aios-orchestrator-gateway.conf"
 install -m 0644 "$SOURCE_ROOT/systemd/aios-orchestrator-remote-tool.service" \
  "$UNIT_ROOT/aios-orchestrator-remote-tool.service"
 if [[ ! -f "$CONFIG_ROOT/inbound.token" ]]; then
  python3 - <<'PY' > "$CONFIG_ROOT/inbound.token"
 import secrets, sys
 sys.stdout.write(secrets.token_urlsafe(48))
 PY
 fi
 [[ ! -L "$CONFIG_ROOT/inbound.token" ]] || fail INBOUND_TOKEN_SYMLINK
 chown root:aios-orchestrator "$CONFIG_ROOT/inbound.token"
 chmod 0640 "$CONFIG_ROOT/inbound.token"
 if [[ ! -f "$CONFIG_ROOT/callback.token" ]]; then
  python3 - <<'PY' > "$CONFIG_ROOT/callback.token"
 import secrets, sys
 sys.stdout.write(secrets.token_urlsafe(48))
 PY
 fi
 [[ ! -L "$CONFIG_ROOT/callback.token" ]] || fail CALLBACK_TOKEN_SYMLINK
 chown root:aios-orchestrator "$CONFIG_ROOT/callback.token"
 chmod 0640 "$CONFIG_ROOT/callback.token"
 # Historical installers appended one LF. Remove only that exact legacy byte in
 # place; reject every other newline form without ever copying or printing data.
 python3 - "$CONFIG_ROOT/config.json" "$CONFIG_ROOT/inbound.token" "$CONFIG_ROOT/callback.token" <<'PY'
 import json, os, pathlib, stat, sys
 config_path = pathlib.Path(sys.argv[1])
 names = list(sys.argv[2:])
 if config_path.is_file() and not config_path.is_symlink():
    config = json.loads(config_path.read_text())
    candidates = [
        config.get("inbound_token_file"),
        config.get("callback_token_file"),
        (config.get("workspace_agent") or {}).get("access_token_file"),
        (config.get("responses_fallback") or {}).get("api_key_file"),
        (config.get("openclaw_sync") or {}).get("bearer_token_file"),
    ]
    names.extend(name for name in candidates if isinstance(name, str) and name)
 for name in dict.fromkeys(names):
    path = pathlib.Path(name)
    if not path.is_absolute() or path.is_symlink() or path.resolve() != path:
        raise SystemExit("invalid token path")
    fd = os.open(name, os.O_RDWR | getattr(os, "O_NOFOLLOW", 0))
    try:
        info = os.fstat(fd)
        if not stat.S_ISREG(info.st_mode) or info.st_size <= 0 or info.st_size > 16384:
            raise SystemExit("invalid token metadata")
        data = os.read(fd, info.st_size)
        if data.endswith(b"\n") and b"\n" not in data[:-1] and b"\r" not in data:
            os.ftruncate(fd, info.st_size - 1)
            os.fsync(fd)
        elif b"\n" in data or b"\r" in data:
            raise SystemExit("invalid token format")
    finally:
        os.close(fd)
 PY
 if [[ ! -f "$CONFIG_ROOT/config.json" ]]; then
  install -m 0600 \
    "$SOURCE_ROOT/config/aios-orchestrator-gateway.example.json" \
    "$CONFIG_ROOT/config.json.example"
  printf 'AIOS_ORCHESTRATOR_INSTALL=BLOCKED reason=CONFIGURATION_REQUIRED digest=%s\n' \
    "$SOURCE_DIGEST"
  exit 2
 fi
 chown root:aios-orchestrator "$CONFIG_ROOT/config.json"
 chmod 0640 "$CONFIG_ROOT/config.json"
 python3 -m json.tool "$CONFIG_ROOT/config.json" >/dev/null
 AIOS_ORCHESTRATOR_CONFIG_OWNER_UID=0 \
 AIOS_ORCHESTRATOR_SECRET_OWNER_UID=0 \
 PYTHONPATH="$RELEASE_DIR" \
 python3 - "$CONFIG_ROOT/config.json" <<'PY'
 import sys
 from pathlib import Path
 from services.aios_orchestrator_gateway.config import Config
 cfg = Config.from_file(Path(sys.argv[1]))
 cfg.inbound_token()
 cfg.callback_token()
 if cfg.workspace_trigger_id:
    cfg.workspace_token()
 if cfg.responses_api_key_file:
    cfg.responses_api_key()
 if cfg.openclaw_sync_endpoint:
    cfg.openclaw_sync_token()
 print("CONFIG_VALIDATE=PASS")
 PY
 # Commit the release pointer only after every candidate/config/secret check has
 # succeeded. A failed validation therefore leaves the running release intact.
 ln -s "$RELEASE_DIR" "$INSTALL_ROOT/.current-new-$$"
 mv -Tf "$INSTALL_ROOT/.current-new-$$" "$INSTALL_ROOT/current"
 python3 - "$STATE_ROOT/install-receipt.json" "$SOURCE_DIGEST" "$RELEASE_DIR" "$PREVIOUS_TARGET" <<'PY'
 import json, os, sys, tempfile
 from pathlib import Path
 path = Path(sys.argv[1])
 data = {
    "schema_version": 1,
    "source_digest": sys.argv[2],
    "current_release": sys.argv[3],
    "previous_release": sys.argv[4] or None,
    "last_rollback": None,
 }
 path.parent.mkdir(parents=True, exist_ok=True)
 fd, temp = tempfile.mkstemp(prefix=".install-receipt-", dir=path.parent)
 try:
    with os.fdopen(fd, "w") as handle:
        json.dump(data, handle, indent=2, sort_keys=True)
        handle.write("\n")
        handle.flush()
        os.fsync(handle.fileno())
    os.chmod(temp, 0o600)
    os.replace(temp, path)
    directory = os.open(path.parent, os.O_RDONLY | os.O_DIRECTORY)
    try:
        os.fsync(directory)
    finally:
        os.close(directory)
 finally:
    try:
        os.unlink(temp)
    except FileNotFoundError:
        pass
 PY
 chown root:root "$STATE_ROOT/install-receipt.json"
 chmod 0600 "$STATE_ROOT/install-receipt.json"
 systemctl daemon-reload
 systemctl enable aios-orchestrator-gateway.service
 systemctl enable aios-orchestrator-gateway-poll.timer
 systemctl stop aios-orchestrator-gateway-poll.timer || true
 systemctl stop aios-orchestrator-gateway-poll.service || true
 rollback_activation() {
  if [[ -n "$PREVIOUS_TARGET" && -d "$PREVIOUS_TARGET" ]]; then
    bash "$RELEASE_DIR/scripts/rollback_aios_orchestrator_gateway.sh" || true
  else
    systemctl stop aios-orchestrator-gateway.service || true
  fi
  fail ACTIVATION_FAILED_ROLLED_BACK
 }
 systemctl restart aios-orchestrator-gateway.service || rollback_activation
 python3 - <<'PY' || rollback_activation
 import json, time, urllib.request
 for _ in range(30):
    try:
        with urllib.request.urlopen("http://127.0.0.1:18920/healthz", timeout=1) as response:
            body = json.loads(response.read(65536))
            if response.status == 200 and body.get("status") == "active":
                raise SystemExit(0)
    except Exception:
        pass
    time.sleep(0.5)
 raise SystemExit(1)
 PY
 systemctl start aios-orchestrator-gateway-poll.timer || rollback_activation
 systemctl is-active --quiet aios-orchestrator-gateway.service
 systemctl is-active --quiet aios-orchestrator-gateway-poll.timer
 AIOS_ORCHESTRATOR_CONFIG_OWNER_UID=0 \
 AIOS_ORCHESTRATOR_SECRET_OWNER_UID=0 \
 PYTHONPATH="$INSTALL_ROOT/current" \
 python3 "$INSTALL_ROOT/current/scripts/bootstrap_orchestration_loop.py" \
  --config "$CONFIG_ROOT/config.json"
 systemctl start aios-orchestrator-gateway-poll.service || rollback_activation
 if [[ -f "$CONFIG_ROOT/remote-tool.json" ]]; then
  [[ ! -L "$CONFIG_ROOT/remote-tool.json" ]] || fail REMOTE_TOOL_CONFIG_SYMLINK
  chown root:aios-orchestrator "$CONFIG_ROOT/remote-tool.json"
  chmod 0640 "$CONFIG_ROOT/remote-tool.json"
  PYTHONPATH="$RELEASE_DIR" python3 - "$CONFIG_ROOT/remote-tool.json" <<'PY'
 import sys
 from pathlib import Path
 from services.aios_orchestrator_gateway.remote_tool_server import RemoteToolServerConfig
 RemoteToolServerConfig.from_file(Path(sys.argv[1]))
 print("REMOTE_TOOL_CONFIG_VALIDATE=PASS")
 PY
  PYTHONPATH="$RELEASE_DIR" python3 \
    "$RELEASE_DIR/scripts/generate_orchestrator_tool_manifest.py" \
    --transport-config "$CONFIG_ROOT/remote-tool.json" \
    --output "$STATE_ROOT/orchestrator-tool-manifest.json"
  systemctl enable --now aios-orchestrator-remote-tool.service
 else
  printf 'REMOTE_TOOL_CONFIGURATION=BLOCKED reason=REMOTE_CALLBACK_ENDPOINT_CONFIGURATION\n'
 fi
 printf 'AIOS_ORCHESTRATOR_INSTALL=PASS digest=%s previous=%s\n' \
  "$SOURCE_DIGEST" "${PREVIOUS_TARGET:-NONE}"
@@ -1,112 +0,0 @@
 #!/usr/bin/env bash
 set -euo pipefail
 if [[ ${EUID} -ne 0 ]]; then
  echo "root required" >&2
  exit 1
 fi
 SOURCE_ROOT=${1:?usage: install_automation_quorum_controller.sh SOURCE_ROOT VERSION}
 VERSION=${2:?usage: install_automation_quorum_controller.sh SOURCE_ROOT VERSION}
 [[ ${VERSION} =~ ^[0-9A-Za-z._-]+$ ]] || { echo "invalid version" >&2; exit 1; }
 SOURCE_REAL=$(readlink -f -- "${SOURCE_ROOT}" 2>/dev/null || true)
 [[ ${SOURCE_ROOT} = /* && -n ${SOURCE_REAL} && ${SOURCE_ROOT} == "${SOURCE_REAL}" && -d ${SOURCE_ROOT} && ! -L ${SOURCE_ROOT} ]] || {
  echo "unsafe source root" >&2
  exit 1
 }
 SOURCE_FILES=(
  automation/executor_quorum_controller.py
  automation/health_probe_common.py
  automation/central_gpt_health_probe.py
  automation/codex_executor_health_probe.py
  automation/a0_unattended_loop_probe.py
  schemas/automation-quorum-controller.schema.json
  config/aios-automation-quorum-controller.example.json
  config/aios-health-probes.example.json
  systemd/aios-automation-quorum-controller.service
  systemd/aios-automation-quorum-controller.timer
  systemd/aios-automation-quorum-controller.sysusers
  systemd/aios-automation-quorum-controller.tmpfiles
  systemd/aios-central-gpt-health-probe.service
  systemd/aios-central-gpt-health-probe.timer
  systemd/aios-codex-executor-health-probe.service
  systemd/aios-codex-executor-health-probe.timer
 )
 for relative in "${SOURCE_FILES[@]}"; do
  [[ -f ${SOURCE_ROOT}/${relative} && ! -L ${SOURCE_ROOT}/${relative} ]] || {
    echo "unsafe source file: ${relative}" >&2
    exit 1
  }
 done
 INSTALL_ROOT=/opt/aios-automation-quorum
 VERSION_DIR=${INSTALL_ROOT}/releases/${VERSION}
 [[ ! -e ${VERSION_DIR} ]] || { echo "version already installed" >&2; exit 1; }
 install -d -o root -g root -m 0755 "${VERSION_DIR}/bin" "${VERSION_DIR}/share"
 install -o root -g root -m 0755 "${SOURCE_ROOT}/automation/executor_quorum_controller.py" \
  "${VERSION_DIR}/bin/executor-quorum-controller"
 install -o root -g root -m 0644 "${SOURCE_ROOT}/automation/health_probe_common.py" \
  "${VERSION_DIR}/bin/health_probe_common.py"
 install -o root -g root -m 0755 "${SOURCE_ROOT}/automation/central_gpt_health_probe.py" \
  "${VERSION_DIR}/bin/central-gpt-health-probe"
 install -o root -g root -m 0755 "${SOURCE_ROOT}/automation/codex_executor_health_probe.py" \
  "${VERSION_DIR}/bin/codex-executor-health-probe"
 install -o root -g root -m 0755 "${SOURCE_ROOT}/automation/a0_unattended_loop_probe.py" \
  "${VERSION_DIR}/bin/a0-unattended-loop-probe"
 install -o root -g root -m 0644 "${SOURCE_ROOT}/schemas/automation-quorum-controller.schema.json" \
  "${VERSION_DIR}/share/automation-quorum-controller.schema.json"
 python3 -m py_compile "${VERSION_DIR}/bin/executor-quorum-controller" \
  "${VERSION_DIR}/bin/health_probe_common.py" \
  "${VERSION_DIR}/bin/central-gpt-health-probe" \
  "${VERSION_DIR}/bin/codex-executor-health-probe"
 python3 -m py_compile "${VERSION_DIR}/bin/a0-unattended-loop-probe"
 install -o root -g root -m 0644 "${SOURCE_ROOT}/systemd/aios-automation-quorum-controller.service" /etc/systemd/system/
 install -o root -g root -m 0644 "${SOURCE_ROOT}/systemd/aios-automation-quorum-controller.timer" /etc/systemd/system/
 install -o root -g root -m 0644 "${SOURCE_ROOT}/systemd/aios-central-gpt-health-probe.service" /etc/systemd/system/
 install -o root -g root -m 0644 "${SOURCE_ROOT}/systemd/aios-central-gpt-health-probe.timer" /etc/systemd/system/
 install -o root -g root -m 0644 "${SOURCE_ROOT}/systemd/aios-codex-executor-health-probe.service" /etc/systemd/system/
 install -o root -g root -m 0644 "${SOURCE_ROOT}/systemd/aios-codex-executor-health-probe.timer" /etc/systemd/system/
 install -o root -g root -m 0644 "${SOURCE_ROOT}/systemd/aios-automation-quorum-controller.sysusers" /usr/lib/sysusers.d/aios-automation-quorum-controller.conf
 install -o root -g root -m 0644 "${SOURCE_ROOT}/systemd/aios-automation-quorum-controller.tmpfiles" /usr/lib/tmpfiles.d/aios-automation-quorum-controller.conf
 systemd-sysusers /usr/lib/sysusers.d/aios-automation-quorum-controller.conf
 systemd-tmpfiles --create /usr/lib/tmpfiles.d/aios-automation-quorum-controller.conf
 install -d -o root -g aios-quorum -m 0750 /etc/aios-automation-quorum
 if [[ ! -e /etc/aios-automation-quorum/controller.json ]]; then
  install -o root -g aios-quorum -m 0640 \
    "${SOURCE_ROOT}/config/aios-automation-quorum-controller.example.json" \
    /etc/aios-automation-quorum/controller.json
 fi
 if [[ ! -e /etc/aios-automation-quorum/health-probes.json ]]; then
  install -o root -g root -m 0600 \
    "${SOURCE_ROOT}/config/aios-health-probes.example.json" \
    /etc/aios-automation-quorum/health-probes.json
 fi
 PREVIOUS_TARGET=$(readlink -f "${INSTALL_ROOT}/current" 2>/dev/null || true)
 RECEIPT=${VERSION_DIR}/install-receipt.json
 python3 - "${VERSION_DIR}" "${VERSION}" "${PREVIOUS_TARGET}" "${RECEIPT}" <<'PY'
 import hashlib, json, os, pathlib, sys, tempfile
 root, version, previous, receipt = map(pathlib.Path, (sys.argv[1], sys.argv[2], sys.argv[3] or ".", sys.argv[4]))
 files = {}
 for item in sorted(root.rglob("*")):
    if item.is_file() and not item.is_symlink():
        files[str(item.relative_to(root))] = hashlib.sha256(item.read_bytes()).hexdigest()
 payload = {"version": str(version), "previous_target": "" if str(previous) == "." else str(previous), "files": files}
 fd, temp = tempfile.mkstemp(dir=str(receipt.parent), prefix=".receipt-")
 with os.fdopen(fd, "w", encoding="utf-8") as handle:
    json.dump(payload, handle, sort_keys=True, separators=(",", ":"))
    handle.write("\n")
 os.chmod(temp, 0o600)
 os.replace(temp, receipt)
 PY
 ln -sfn "${VERSION_DIR}" "${INSTALL_ROOT}/.current-${VERSION}"
 mv -Tf "${INSTALL_ROOT}/.current-${VERSION}" "${INSTALL_ROOT}/current"
 systemctl daemon-reload
 echo "installed=${VERSION_DIR}"
 echo "previous=${PREVIOUS_TARGET}"
@@ -1,38 +0,0 @@
 #!/usr/bin/env python3
 from __future__ import annotations
 import argparse
 import json
 from pathlib import Path
 from services.aios_orchestrator_gateway.config import Config
 from services.aios_orchestrator_gateway.recovery import (
    PendingDecisionRecovery,
    RecoverySettings,
 )
 def main() -> int:
    parser = argparse.ArgumentParser()
    parser.add_argument("--config", type=Path, required=True)
    mode = parser.add_mutually_exclusive_group(required=True)
    mode.add_argument("--plan", action="store_true")
    mode.add_argument("--apply-one", action="store_true")
    args = parser.parse_args()
    config = Config.from_file(args.config)
    recovery = PendingDecisionRecovery(RecoverySettings(
        database_path=config.database_path,
        contract_commit=config.contract_commit,
        lease_seconds=config.lease_seconds,
        max_retries=config.max_retries,
        retry_after=config.retry_base_seconds,
    ))
    if args.plan:
        print(json.dumps(recovery.plan(), separators=(",", ":"), sort_keys=True))
    else:
        print(json.dumps({"processed_event_id": recovery.apply_one()}, sort_keys=True))
    return 0
 if __name__ == "__main__":
    raise SystemExit(main())
@@ -1,130 +0,0 @@
 #!/usr/bin/env python3
 """Submit an OpenClaw/Codex task lifecycle event over HTTP or a polled inbox."""
 from __future__ import annotations
 import argparse
 import json
 import os
 from pathlib import Path
 import sys
 import urllib.error
 import urllib.request
 REPO_ROOT = Path(__file__).resolve().parents[1]
 sys.path.insert(0, str(REPO_ROOT))
 from services.aios_orchestrator_gateway.models import Event
 from services.aios_orchestrator_gateway.securefs import atomic_write_bytes_shared, read_secure_json, read_secure_text
 def parser() -> argparse.ArgumentParser:
    p = argparse.ArgumentParser()
    p.add_argument("--project-key", required=True)
    p.add_argument("--contract-commit", required=True)
    p.add_argument("--event-id", required=True)
    p.add_argument("--source-agent", required=True, choices=["openclaw", "codex", "gitea", "system"])
    p.add_argument("--event-type", required=True)
    p.add_argument("--work-unit-id", required=True)
    p.add_argument(
        "--status",
        required=True,
        choices=["READY", "RUNNING", "PASS", "BLOCKED", "FAIL", "TIMEOUT", "NOT_EXECUTED"],
    )
    p.add_argument("--state-version", type=int, required=True)
    p.add_argument("--implementation-commit")
    p.add_argument("--result-commit")
    p.add_argument("--blocker")
    p.add_argument("--evidence-ref", action="append", default=[])
    p.add_argument("--payload-file", type=Path)
    p.add_argument("--endpoint")
    p.add_argument("--token-file", type=Path)
    p.add_argument("--inbox-dir", type=Path)
    p.add_argument("--timeout-seconds", type=int, default=15)
    return p
 def main() -> int:
    args = parser().parse_args()
    if not args.endpoint and not args.inbox_dir:
        raise SystemExit("either --endpoint or --inbox-dir is required")
    if args.endpoint and not args.token_file:
        raise SystemExit("--token-file is required with --endpoint")
    payload = {}
    if args.payload_file:
        if not args.payload_file.is_absolute():
            raise SystemExit("--payload-file must be absolute")
        payload = read_secure_json(
            args.payload_file, expected_uid=os.getuid(), max_bytes=262_144
        )
        if not isinstance(payload, dict):
            raise SystemExit("payload root must be an object")
    event = Event.from_dict(
        {
            "schema_version": 1,
            "project_key": args.project_key,
            "contract_commit": args.contract_commit,
            "event_id": args.event_id,
            "source_agent": args.source_agent,
            "event_type": args.event_type,
            "work_unit_id": args.work_unit_id,
            "status": args.status,
            "state_version": args.state_version,
            "implementation_commit": args.implementation_commit,
            "result_commit": args.result_commit,
            "blocker": args.blocker,
            "evidence_refs": args.evidence_ref,
            "payload": payload,
        }
    )
    encoded = event.canonical_json().encode()
    http_error: Exception | None = None
    if args.endpoint:
        token = read_secure_text(
            args.token_file,
            expected_uid=None,
            max_bytes=64_000,
        )
        request = urllib.request.Request(
            args.endpoint.rstrip("/") + "/v1/orchestrator/events",
            data=encoded,
            method="POST",
            headers={
                "Authorization": f"Bearer {token}",
                "Content-Type": "application/json",
                "Idempotency-Key": event.event_id,
                "User-Agent": "aios-agent-event-reporter/0.2",
            },
        )
        try:
            with urllib.request.urlopen(request, timeout=args.timeout_seconds) as response:
                response.read(65_536)
                if response.status not in {200, 202}:
                    raise RuntimeError(f"unexpected HTTP {response.status}")
                print(f"EVENT_SUBMIT=PASS event_id={event.event_id} channel=http")
                return 0
        except (urllib.error.URLError, urllib.error.HTTPError, TimeoutError, OSError) as exc:
            http_error = exc
    if args.inbox_dir:
        pending = args.inbox_dir / "pending"
        path = pending / f"{event.event_id}.json"
        atomic_write_bytes_shared(
            path,
            (json.dumps(event.to_dict(), indent=2, sort_keys=True) + "\n").encode(),
            mode=0o640,
            group_writable_directory=True,
        )
        print(f"EVENT_SUBMIT=PASS event_id={event.event_id} channel=file_inbox")
        return 0
    print(
        f"EVENT_SUBMIT=FAIL event_id={event.event_id} reason={type(http_error).__name__}",
        file=sys.stderr,
    )
    return 2
 if __name__ == "__main__":
    raise SystemExit(main())
@@ -1,102 +0,0 @@
 #!/usr/bin/env bash
 set -Eeuo pipefail
 umask 0077
 INSTALL_ROOT="${INSTALL_ROOT:-/opt/aios-orchestrator-gateway}"
 STATE_ROOT="${STATE_ROOT:-/var/lib/aios-orchestrator-gateway}"
 RECEIPT="$STATE_ROOT/install-receipt.json"
 fail() {
  printf 'AIOS_ORCHESTRATOR_ROLLBACK=BLOCKED reason=%s\n' "$1" >&2
  exit 2
 }
 [[ "$(id -u)" -eq 0 ]] || fail OWNER_REQUIRED
 [[ -f "$RECEIPT" ]] || fail RECEIPT_MISSING
 [[ ! -L "$RECEIPT" ]] || fail RECEIPT_SYMLINK
 [[ "$(stat -c '%u:%a' "$RECEIPT")" == "0:600" ]] || fail RECEIPT_PERMISSION
 readarray -t VALUES < <(
 python3 - "$RECEIPT" "$INSTALL_ROOT/releases" "$INSTALL_ROOT/current" <<'PY'
 import json, os, stat, sys
 from pathlib import Path
 receipt = Path(sys.argv[1])
 releases = Path(sys.argv[2]).resolve()
 current_link = Path(sys.argv[3])
 data = json.loads(receipt.read_text())
 previous = data.get("previous_release")
 current = data.get("current_release")
 if not previous or not current:
    raise SystemExit(2)
 previous_path = Path(previous).resolve()
 current_path = Path(current).resolve()
 if previous_path.parent != releases or current_path.parent != releases:
    raise SystemExit(2)
 if not stat.S_ISDIR(os.lstat(previous_path).st_mode) or not stat.S_ISDIR(os.lstat(current_path).st_mode):
    raise SystemExit(2)
 for release in (previous_path, current_path):
    if len(release.name) != 64 or any(ch not in "0123456789abcdef" for ch in release.name):
        raise SystemExit(2)
    digest_file = release / "SOURCE_DIGEST"
    if digest_file.is_symlink() or not digest_file.is_file():
        raise SystemExit(2)
    if digest_file.read_text(encoding="utf-8").strip() != release.name:
        raise SystemExit(2)
 last = data.get("last_rollback")
 if isinstance(last, dict) and last.get("to") == current and current_link.is_symlink():
    if current_link.resolve() == current_path:
        print("ALREADY_ROLLED_BACK")
        raise SystemExit(0)
 print(previous_path)
 print(current_path)
 PY
 ) || fail RECEIPT_INVALID
 if [[ "${VALUES[0]:-}" == "ALREADY_ROLLED_BACK" ]]; then
  echo "AIOS_ORCHESTRATOR_ROLLBACK=PASS result=NOOP_ALREADY_ROLLED_BACK"
  exit 0
 fi
 [[ "${#VALUES[@]}" -eq 2 ]] || fail RECEIPT_INVALID
 PREVIOUS="${VALUES[0]}"
 CURRENT="${VALUES[1]}"
 systemctl stop aios-orchestrator-gateway-poll.timer || true
 systemctl stop aios-orchestrator-gateway-poll.service || true
 systemctl stop aios-orchestrator-gateway.service || true
 ln -s "$PREVIOUS" "$INSTALL_ROOT/.current-rollback-$$"
 mv -Tf "$INSTALL_ROOT/.current-rollback-$$" "$INSTALL_ROOT/current"
 systemctl start aios-orchestrator-gateway.service
 systemctl is-active --quiet aios-orchestrator-gateway.service
 systemctl start aios-orchestrator-gateway-poll.timer
 systemctl is-active --quiet aios-orchestrator-gateway-poll.timer
 python3 - "$RECEIPT" "$PREVIOUS" "$CURRENT" <<'PY'
 import json, os, sys, tempfile
 from pathlib import Path
 path = Path(sys.argv[1])
 data = json.loads(path.read_text())
 data["current_release"] = sys.argv[2]
 data["previous_release"] = sys.argv[3]
 data["last_rollback"] = {"from": sys.argv[3], "to": sys.argv[2]}
 fd, temp = tempfile.mkstemp(prefix=".rollback-receipt-", dir=path.parent)
 try:
    with os.fdopen(fd, "w") as handle:
        json.dump(data, handle, indent=2, sort_keys=True)
        handle.write("\n")
        handle.flush()
        os.fsync(handle.fileno())
    os.chmod(temp, 0o600)
    os.chown(temp, 0, 0)
    os.replace(temp, path)
    directory = os.open(path.parent, os.O_RDONLY | os.O_DIRECTORY)
    try:
        os.fsync(directory)
    finally:
        os.close(directory)
 finally:
    try:
        os.unlink(temp)
    except FileNotFoundError:
        pass
 PY
 printf 'AIOS_ORCHESTRATOR_ROLLBACK=PASS current=%s previous=%s\n' "$PREVIOUS" "$CURRENT"
@@ -1,29 +0,0 @@
 #!/usr/bin/env bash
 set -euo pipefail
 if [[ ${EUID} -ne 0 ]]; then
  echo "root required" >&2
  exit 1
 fi
 TARGET=${1:?usage: rollback_automation_quorum_controller.sh ABSOLUTE_RELEASE_PATH}
 INSTALL_ROOT=/opt/aios-automation-quorum
 TARGET_REAL=$(readlink -f -- "${TARGET}" 2>/dev/null || true)
 [[ -n ${TARGET_REAL} && ${TARGET} == "${TARGET_REAL}" ]] || { echo "non-canonical or missing rollback target" >&2; exit 1; }
 case ${TARGET_REAL} in
  "${INSTALL_ROOT}"/releases/*) ;;
  *) echo "target outside release root" >&2; exit 1 ;;
 esac
 [[ -d ${TARGET_REAL} && ! -L ${TARGET} ]] || { echo "invalid rollback target" >&2; exit 1; }
 CURRENT=$(readlink -f "${INSTALL_ROOT}/current" 2>/dev/null || true)
 if [[ ${CURRENT} == "${TARGET_REAL}" ]]; then
  echo "already_current=${CURRENT}"
  exit 0
 fi
 ln -sfn "${TARGET_REAL}" "${INSTALL_ROOT}/.rollback-current"
 mv -Tf "${INSTALL_ROOT}/.rollback-current" "${INSTALL_ROOT}/current"
 systemctl daemon-reload
 systemctl try-restart aios-automation-quorum-controller.timer
 echo "rolled_back_to=$(readlink -f "${INSTALL_ROOT}/current")"
@@ -1,117 +0,0 @@
 #!/usr/bin/env python3
 """Dependency-free, deterministic validation for the shipped controller config."""
 from __future__ import annotations
 import json
 import sys
 from pathlib import Path
 from urllib.parse import urlparse
 REQUIRED = {
    "project_key",
    "contract_commit",
    "database_path",
    "file_inbox_path",
    "health_receipts",
    "blocker_retry_threshold",
    "max_health_age_seconds",
    "lease_ttl_seconds",
    "gateway_event_url",
    "gateway_token_file",
 }
 OPTIONAL = {"$schema", "default_workload", "git_ref_polling"}
 WORKLOADS = {
    "large_repetitive_infrastructure_monitoring",
    "complex_code_security_architecture",
 }
 def fail(message: str) -> None:
    raise ValueError(message)
 def absolute_path(value: object, name: str) -> None:
    if not isinstance(value, str) or not value.startswith("/") or "\x00" in value:
        fail(f"{name} must be an absolute POSIX path")
 def bounded_integer(value: object, name: str, minimum: int, maximum: int) -> None:
    if isinstance(value, bool) or not isinstance(value, int) or not minimum <= value <= maximum:
        fail(f"{name} must be an integer in [{minimum}, {maximum}]")
 def validate(config: object, schema: object) -> None:
    if not isinstance(config, dict) or not isinstance(schema, dict):
        fail("config and schema must be JSON objects")
    if schema.get("$schema") != "https://json-schema.org/draft/2020-12/schema":
        fail("unexpected JSON Schema dialect")
    if schema.get("additionalProperties") is not False:
        fail("schema must reject additional properties")
    if set(schema.get("required", [])) != REQUIRED:
        fail("schema required fields drifted from validator")
    keys = set(config)
    if not REQUIRED <= keys or keys - REQUIRED - OPTIONAL:
        fail("missing or unexpected top-level config fields")
    absolute_path(config["database_path"], "database_path")
    absolute_path(config["file_inbox_path"], "file_inbox_path")
    absolute_path(config["gateway_token_file"], "gateway_token_file")
    project_key = config["project_key"]
    if not isinstance(project_key, str) or not project_key or len(project_key) > 128:
        fail("project_key is invalid")
    commit = config["contract_commit"]
    if not isinstance(commit, str) or len(commit) != 40 or any(c not in "0123456789abcdef" for c in commit):
        fail("contract_commit must be a full lowercase Git OID")
    receipts = config["health_receipts"]
    if not isinstance(receipts, dict) or set(receipts) != {"central_gpt", "openclaw", "codex"}:
        fail("health_receipts must contain exactly central_gpt, openclaw, and codex")
    for subject, value in receipts.items():
        absolute_path(value, f"health_receipts.{subject}")
    git_polling = config.get("git_ref_polling")
    if git_polling is not None:
        if not isinstance(git_polling, dict) or set(git_polling) != {"repository", "refs"}:
            fail("git_ref_polling must contain exactly repository and refs")
        absolute_path(git_polling["repository"], "git_ref_polling.repository")
        refs = git_polling["refs"]
        if not isinstance(refs, list) or not refs or len(refs) != len(set(refs)):
            fail("git_ref_polling.refs must be a non-empty unique list")
        if any(not isinstance(ref, str) or not ref.startswith("refs/") or ".." in ref for ref in refs):
            fail("git_ref_polling.refs contains an invalid ref")
    if config.get("default_workload", "large_repetitive_infrastructure_monitoring") not in WORKLOADS:
        fail("unsupported default_workload")
    bounded_integer(config["blocker_retry_threshold"], "blocker_retry_threshold", 1, 20)
    bounded_integer(config["max_health_age_seconds"], "max_health_age_seconds", 5, 300)
    bounded_integer(config["lease_ttl_seconds"], "lease_ttl_seconds", 15, 3600)
    url = config["gateway_event_url"]
    parsed = urlparse(url) if isinstance(url, str) else None
    if (
        not parsed
        or parsed.scheme != "http"
        or parsed.hostname != "127.0.0.1"
        or parsed.port != 18920
        or parsed.path != "/v1/orchestrator/events"
        or parsed.query
        or parsed.fragment
    ):
        fail("gateway_event_url must be the fixed loopback orchestrator endpoint")
 def main() -> int:
    if len(sys.argv) != 3:
        print(f"usage: {Path(sys.argv[0]).name} CONFIG SCHEMA", file=sys.stderr)
        return 2
    with Path(sys.argv[1]).open("r", encoding="utf-8") as handle:
        config = json.load(handle)
    with Path(sys.argv[2]).open("r", encoding="utf-8") as handle:
        schema = json.load(handle)
    try:
        validate(config, schema)
    except ValueError as error:
        print(f"schema validation failed: {error}", file=sys.stderr)
        return 1
    return 0
 if __name__ == "__main__":
    raise SystemExit(main())
@@ -1,32 +0,0 @@
 #!/usr/bin/env bash
 set -Eeuo pipefail
 CONFIG="${1:-/etc/aios-orchestrator-gateway/config.json}"
 PORT="${PORT:-18920}"
 python3 -m json.tool "$CONFIG" >/dev/null
 systemd-analyze verify \
  /etc/systemd/system/aios-orchestrator-gateway.service \
  /etc/systemd/system/aios-orchestrator-gateway-poll.service \
  /etc/systemd/system/aios-orchestrator-gateway-poll.timer
 systemctl is-active --quiet aios-orchestrator-gateway.service
 systemctl is-active --quiet aios-orchestrator-gateway-poll.timer
 health="$(curl --fail --silent --show-error "http://127.0.0.1:${PORT}/healthz")"
 python3 -c '
 import json, sys
 v=json.load(sys.stdin)
 assert v["service"]=="aios-orchestrator-gateway"
 assert v["status"]=="active"
 assert v["polling"] is True
 assert len(v["contract_commit"])==40
 ' <<<"$health"
 systemctl start aios-orchestrator-gateway-poll.service
 if journalctl -u aios-orchestrator-gateway.service -n 200 --no-pager \
  | grep -Eqi 'Bearer |token=|api[_-]?key|secret='; then
  echo "AIOS_ORCHESTRATOR_VERIFY=BLOCKED reason=SECRET_PATTERN_IN_LOG" >&2
  exit 2
 fi
 echo "AIOS_ORCHESTRATOR_VERIFY=PASS"
@@ -1,29 +0,0 @@
 #!/usr/bin/env bash
 set -euo pipefail
 ROOT=${1:-$(cd "$(dirname "${BASH_SOURCE[0]}")/.." && pwd)}
 python3 -m py_compile "${ROOT}/automation/executor_quorum_controller.py"
 python3 -m py_compile \
  "${ROOT}/automation/health_probe_common.py" \
  "${ROOT}/automation/central_gpt_health_probe.py" \
  "${ROOT}/automation/codex_executor_health_probe.py" \
  "${ROOT}/automation/a0_unattended_loop_probe.py"
 python3 -m unittest discover -s "${ROOT}/tests" -p 'test_executor_quorum_controller.py' -v
 python3 -m unittest discover -s "${ROOT}/tests" -p 'test_aios_health_probes.py' -v
 python3 -m json.tool "${ROOT}/config/aios-automation-quorum-controller.example.json" >/dev/null
 python3 -m json.tool "${ROOT}/schemas/automation-quorum-controller.schema.json" >/dev/null
 python3 "${ROOT}/scripts/validate_automation_quorum_schema.py" \
  "${ROOT}/config/aios-automation-quorum-controller.example.json" \
  "${ROOT}/schemas/automation-quorum-controller.schema.json"
 python3 -m py_compile "${ROOT}/scripts/validate_automation_quorum_schema.py"
 bash -n "${ROOT}/scripts/install_automation_quorum_controller.sh"
 bash -n "${ROOT}/scripts/rollback_automation_quorum_controller.sh"
 bash -n "${ROOT}/scripts/verify_automation_quorum_controller.sh"
 systemd-analyze verify \
  "${ROOT}/systemd/aios-automation-quorum-controller.service" \
  "${ROOT}/systemd/aios-automation-quorum-controller.timer" \
  "${ROOT}/systemd/aios-central-gpt-health-probe.service" \
  "${ROOT}/systemd/aios-central-gpt-health-probe.timer" \
  "${ROOT}/systemd/aios-codex-executor-health-probe.service" \
  "${ROOT}/systemd/aios-codex-executor-health-probe.timer"
@@ -1,652 +0,0 @@
 #!/usr/bin/env python3
 """Hardened WeChat outbound backend for AIOS.
 Only the dedicated UCC adapter is a supported send backend.  OpenClaw channel
 discovery is retained for diagnostics, but generic OpenClaw chat or agent
 delivery is never promoted to WeChat delivery.
 """
 from __future__ import annotations
 import json
 import os
 import re
 import selectors
 import signal
 import subprocess
 import time
 import urllib.error
 import urllib.parse
 import urllib.request
 from typing import Any, Dict, List, Optional, Tuple
 OPENCLAW_BIN = os.environ.get("AIOS_OPENCLAW_BIN", "openclaw")
 WECHAT_ACCOUNT_ID = os.environ.get("AIOS_WECHAT_ACCOUNT_ID", "")
 WECHAT_CHANNEL = os.environ.get("AIOS_WECHAT_CHANNEL", "")
 SEND_BACKEND = os.environ.get("AIOS_WECHAT_SEND_BACKEND", "")
 UCC_ADAPTER_URL = os.environ.get("AIOS_UCC_WECHAT_ADAPTER_URL", "http://127.0.0.1:18800")
 UCC_ADAPTER_TOKEN = os.environ.get("AIOS_UCC_WECHAT_ADAPTER_TOKEN") or os.environ.get(
    "AIOS_WECHAT_QR_ADAPTER_TOKEN", ""
 )
 UCC_SEND_PATH = os.environ.get("AIOS_UCC_WECHAT_SEND_PATH", "/api/internal/wechat/send")
 _allowed_origins = os.environ.get("AIOS_UCC_WECHAT_ALLOWED_ORIGINS", "").strip()
 UCC_ADAPTER_ALLOWED_ORIGINS = tuple(
    origin.strip() for origin in _allowed_origins.split(",") if origin.strip()
 ) or (
    "http://127.0.0.1:18800",
    "http://[::1]:18800",
 )
 MAX_TARGET_BYTES = 512
 MAX_CONTENT_BYTES = 32 * 1024
 MAX_IDENTIFIER_BYTES = 512
 MAX_METADATA_BYTES = 8 * 1024
 MAX_METADATA_DEPTH = 32
 MAX_METADATA_NODES = 1024
 MAX_RESPONSE_BYTES = 64 * 1024
 MAX_ENDPOINT_BYTES = 256
 _DEFAULT_TIMEOUT_SECONDS = 10
 _MAX_TIMEOUT_SECONDS = 30
 _LEGACY_SEND_PATH = "/internal/channels/wechat/send"
 _SAFE_PATH = re.compile(r"^/[A-Za-z0-9._~!$&'()*+,;=:@%/-]+$")
 _MAX_PATH_DECODE_ROUNDS = 4
 _OPENCLAW_TIMEOUT_RETURN_CODE = 124
 _OPENCLAW_OUTPUT_LIMIT_RETURN_CODE = 125
 _OPENCLAW_SPAWN_ERROR_RETURN_CODE = 126
 def _bounded_timeout(value: str) -> int:
    try:
        parsed = int(value)
    except (TypeError, ValueError):
        return _DEFAULT_TIMEOUT_SECONDS
    return min(max(parsed, 1), _MAX_TIMEOUT_SECONDS)
 REQUEST_TIMEOUT_SECONDS = _bounded_timeout(os.environ.get("AIOS_UCC_WECHAT_TIMEOUT_SECONDS", "10"))
 class BackendResult:
    def __init__(self, ok: bool, backend: str, status: str, data: Dict[str, Any]) -> None:
        self.ok = ok
        self.backend = backend
        self.status = status
        self.data = data
    def to_dict(self) -> Dict[str, Any]:
        return {"ok": self.ok, "backend": self.backend, "status": self.status, **self.data}
 class _BackendProtocolError(Exception):
    """Internal marker whose details must never cross the module boundary."""
 def _openclaw_base_args() -> List[str]:
    """Return the OpenClaw 2026.6.11-compatible command prefix.
    Gateway/auth global flags are not supported by the installed CLI and would
    also expose credentials through the process argument vector.
    """
    return [OPENCLAW_BIN]
 def _terminate_process_tree(process: Any) -> None:
    if os.name == "posix":
        try:
            os.killpg(process.pid, signal.SIGTERM)
        except OSError:
            pass
        try:
            process.wait(timeout=0.5)
        except (OSError, subprocess.TimeoutExpired):
            pass
        time.sleep(0.05)
        try:
            os.killpg(process.pid, signal.SIGKILL)
        except OSError:
            pass
        try:
            process.wait(timeout=0.5)
        except (OSError, subprocess.TimeoutExpired):
            pass
        return
    if process.poll() is not None:
        return
    try:
        process.terminate()
        process.wait(timeout=1)
        return
    except (OSError, subprocess.TimeoutExpired):
        pass
    try:
        process.kill()
        process.wait(timeout=1)
    except (OSError, subprocess.TimeoutExpired):
        pass
 def _run_openclaw(
    args: List[str], timeout: int = 15
 ) -> subprocess.CompletedProcess[str]:
    bounded_timeout = min(max(int(timeout), 1), _MAX_TIMEOUT_SECONDS)
    popen_options: Dict[str, Any] = {
        "stdout": subprocess.PIPE,
        "stderr": subprocess.PIPE,
    }
    if os.name == "posix":
        popen_options["start_new_session"] = True
    elif hasattr(subprocess, "CREATE_NEW_PROCESS_GROUP"):
        popen_options["creationflags"] = subprocess.CREATE_NEW_PROCESS_GROUP
    try:
        process = subprocess.Popen(args, **popen_options)
    except (OSError, ValueError):
        return subprocess.CompletedProcess(args, _OPENCLAW_SPAWN_ERROR_RETURN_CODE, "", "")
    selector = selectors.DefaultSelector()
    streams = {"stdout": process.stdout, "stderr": process.stderr}
    buffers = {"stdout": bytearray(), "stderr": bytearray()}
    try:
        for name, stream in streams.items():
            if stream is not None:
                selector.register(stream, selectors.EVENT_READ, name)
    except (OSError, ValueError):
        selector.close()
        _terminate_process_tree(process)
        for stream in streams.values():
            if stream is not None:
                stream.close()
        return subprocess.CompletedProcess(args, _OPENCLAW_SPAWN_ERROR_RETURN_CODE, "", "")
    deadline = time.monotonic() + bounded_timeout
    total_bytes = 0
    return_code: Optional[int] = None
    try:
        while selector.get_map():
            remaining = deadline - time.monotonic()
            if remaining <= 0:
                return_code = _OPENCLAW_TIMEOUT_RETURN_CODE
                break
            events = selector.select(min(remaining, 0.1))
            for key, _mask in events:
                allowance = MAX_RESPONSE_BYTES - total_bytes
                try:
                    chunk = os.read(key.fileobj.fileno(), min(8192, allowance + 1))
                except OSError:
                    return_code = _OPENCLAW_SPAWN_ERROR_RETURN_CODE
                    break
                if not chunk:
                    selector.unregister(key.fileobj)
                    key.fileobj.close()
                    continue
                if len(chunk) > allowance:
                    return_code = _OPENCLAW_OUTPUT_LIMIT_RETURN_CODE
                    break
                buffers[key.data].extend(chunk)
                total_bytes += len(chunk)
            if return_code is not None:
                break
    finally:
        selector.close()
    if return_code is not None:
        _terminate_process_tree(process)
        for stream in streams.values():
            if stream is not None and not stream.closed:
                stream.close()
        return subprocess.CompletedProcess(args, return_code, "", "")
    try:
        completed_return_code = process.wait(timeout=1)
    except subprocess.TimeoutExpired:
        _terminate_process_tree(process)
        return subprocess.CompletedProcess(args, _OPENCLAW_TIMEOUT_RETURN_CODE, "", "")
    stdout = buffers["stdout"].decode("utf-8", errors="replace")
    stderr = buffers["stderr"].decode("utf-8", errors="replace")
    return subprocess.CompletedProcess(args, completed_return_code, stdout, stderr)
 def _json_dict(text: str) -> Dict[str, Any]:
    try:
        value = json.loads(text or "{}")
    except (RecursionError, TypeError, ValueError, json.JSONDecodeError):
        return {}
    return value if isinstance(value, dict) else {}
 def discover_openclaw_wechat_channel() -> Optional[str]:
    """Discover an explicitly typed WeChat/Weixin channel for diagnostics."""
    if WECHAT_CHANNEL:
        return WECHAT_CHANNEL
    try:
        proc = _run_openclaw(_openclaw_base_args() + ["channels", "list", "--json"], timeout=15)
    except (OSError, subprocess.SubprocessError, ValueError):
        return None
    if proc.returncode != 0:
        return None
    data = _json_dict(proc.stdout[:MAX_RESPONSE_BYTES])
    candidates: List[str] = []
    def walk(value: Any) -> None:
        if isinstance(value, dict):
            name = value.get("name") or value.get("channel") or value.get("id")
            kind = value.get("type") or value.get("kind")
            if isinstance(name, str) and isinstance(kind, str):
                label = f"{name} {kind}".lower()
                if "wechat" in label or "weixin" in label:
                    candidates.append(name)
            for child in value.values():
                walk(child)
        elif isinstance(value, list):
            for child in value:
                walk(child)
    walk(data)
    return candidates[0] if candidates else None
 def _auth_headers() -> Dict[str, str]:
    headers = {"Content-Type": "application/json", "Accept": "application/json"}
    if UCC_ADAPTER_TOKEN:
        headers["Authorization"] = "Bearer " + UCC_ADAPTER_TOKEN
    return headers
 def _safe_path(value: Any) -> Optional[str]:
    if not isinstance(value, str):
        return None
    path = value if value.startswith("/") else "/" + value
    try:
        encoded_path = path.encode("utf-8")
    except UnicodeEncodeError:
        return None
    if len(encoded_path) > MAX_ENDPOINT_BYTES:
        return None
    if not _SAFE_PATH.fullmatch(path):
        return None
    decoded = path
    for _ in range(_MAX_PATH_DECODE_ROUNDS):
        try:
            next_decoded = urllib.parse.unquote_to_bytes(decoded).decode("utf-8")
        except UnicodeDecodeError:
            return None
        if next_decoded == decoded:
            break
        decoded = next_decoded
    if "%" in decoded or "\\" in decoded or "\x00" in decoded:
        return None
    if not _SAFE_PATH.fullmatch(decoded):
        return None
    if any(segment in {".", ".."} for segment in decoded.split("/")):
        return None
    return path
 def _unique_paths(paths: List[str]) -> List[str]:
    out: List[str] = []
    for candidate in paths:
        path = _safe_path(candidate)
        if path and path not in out:
            out.append(path)
        if len(out) == 2:
            break
    return out
 def _ucc_send_paths() -> List[str]:
    canonical = _safe_path(UCC_SEND_PATH) or "/api/internal/wechat/send"
    return _unique_paths([canonical, _LEGACY_SEND_PATH])
 def _canonical_origin(value: Any) -> Optional[str]:
    if not isinstance(value, str) or not value or len(value) > 2048:
        return None
    if "\\" in value or any(ord(character) < 32 for character in value):
        return None
    try:
        parsed = urllib.parse.urlsplit(value)
        port = parsed.port
    except (UnicodeError, ValueError):
        return None
    scheme = parsed.scheme.lower()
    if scheme not in {"http", "https"} or not parsed.hostname:
        return None
    if parsed.username or parsed.password or parsed.query or parsed.fragment or parsed.path not in {"", "/"}:
        return None
    try:
        host = parsed.hostname.rstrip(".").encode("idna").decode("ascii").lower()
    except UnicodeError:
        return None
    if not host:
        return None
    if port is None:
        port = 80 if scheme == "http" else 443
    rendered_host = f"[{host}]" if ":" in host else host
    return f"{scheme}://{rendered_host}:{port}"
 def _adapter_base_url() -> Optional[str]:
    origin = _canonical_origin(UCC_ADAPTER_URL)
    allowed = {
        canonical
        for canonical in (_canonical_origin(value) for value in UCC_ADAPTER_ALLOWED_ORIGINS)
        if canonical
    }
    return origin if origin and origin in allowed else None
 class _NoRedirectHandler(urllib.request.HTTPRedirectHandler):
    def redirect_request(
        self,
        req: Any,
        fp: Any,
        code: int,
        msg: str,
        headers: Any,
        newurl: str,
    ) -> None:
        del req, fp, code, msg, headers, newurl
        return None
 def _open_no_redirect(request: urllib.request.Request, timeout: int) -> Any:
    opener = urllib.request.build_opener(_NoRedirectHandler())
    return opener.open(request, timeout=timeout)
 def ucc_adapter_alive() -> bool:
    base_url = _adapter_base_url()
    if not base_url:
        return False
    request = urllib.request.Request(
        base_url + "/health", headers=_auth_headers(), method="GET"
    )
    try:
        with _open_no_redirect(request, timeout=min(5, REQUEST_TIMEOUT_SECONDS)) as response:
            return 200 <= int(response.status) < 300
    except (OSError, ValueError, urllib.error.URLError):
        return False
 def select_backend() -> Dict[str, Any]:
    explicit = SEND_BACKEND.strip().lower()
    if explicit:
        if explicit == "ucc-adapter":
            return {"backend": "ucc-adapter", "reason": "explicit"}
        return {"backend": "unavailable", "reason": "unsupported_backend"}
    if ucc_adapter_alive():
        return {"backend": "ucc-adapter", "reason": "ucc_adapter_alive"}
    return {"backend": "unavailable", "reason": "ucc_adapter_unavailable"}
 def _post_ucc(path: str, payload: Dict[str, Any]) -> Tuple[int, Dict[str, Any]]:
    base_url = _adapter_base_url()
    safe_path = _safe_path(path)
    if not base_url or not safe_path:
        raise _BackendProtocolError("invalid adapter endpoint")
    try:
        body = json.dumps(payload, ensure_ascii=False, separators=(",", ":")).encode("utf-8")
    except (TypeError, ValueError) as exc:
        raise _BackendProtocolError("invalid request payload") from exc
    request = urllib.request.Request(
        base_url + safe_path,
        data=body,
        headers=_auth_headers(),
        method="POST",
    )
    with _open_no_redirect(request, timeout=REQUEST_TIMEOUT_SECONDS) as response:
        raw = response.read(MAX_RESPONSE_BYTES + 1)
        if len(raw) > MAX_RESPONSE_BYTES:
            raise _BackendProtocolError("adapter response too large")
        try:
            text = raw.decode("utf-8")
        except UnicodeDecodeError as exc:
            raise _BackendProtocolError("adapter response encoding invalid") from exc
        data = _json_dict(text)
        if not data:
            raise _BackendProtocolError("adapter response invalid")
        return int(response.status), data
 def _ucc_response_sent(data: Dict[str, Any]) -> bool:
    return isinstance(data, dict) and data.get("ok") is True and data.get("status") == "sent"
 def _ucc_result_is_ok(data: Dict[str, Any], dry_run: bool = False) -> bool:
    expected_status = "dry_run" if dry_run else "sent"
    return isinstance(data, dict) and data.get("ok") is True and data.get("status") == expected_status
 def _utf8_size(value: str) -> Optional[int]:
    try:
        return len(value.encode("utf-8"))
    except UnicodeEncodeError:
        return None
 def _valid_identifier(value: Any, required: bool = False) -> bool:
    if not isinstance(value, str):
        return False
    if required and not value.strip():
        return False
    if not value:
        return not required
    if len(value) > MAX_IDENTIFIER_BYTES:
        return False
    size = _utf8_size(value)
    return size is not None and size <= MAX_IDENTIFIER_BYTES and not any(
        ord(character) < 32 or ord(character) == 127 for character in value
    )
 def _validate_metadata(metadata: Any) -> Optional[str]:
    if not isinstance(metadata, dict):
        return "invalid_input"
    nodes = 0
    stack: List[Tuple[Any, int]] = [(metadata, 0)]
    while stack:
        value, depth = stack.pop()
        nodes += 1
        if depth > MAX_METADATA_DEPTH or nodes > MAX_METADATA_NODES:
            return "input_too_large"
        if isinstance(value, dict):
            if len(value) + nodes + len(stack) > MAX_METADATA_NODES:
                return "input_too_large"
            for key, child in value.items():
                if (
                    not isinstance(key, str)
                    or len(key) > MAX_METADATA_BYTES
                    or _utf8_size(key) is None
                ):
                    return "invalid_input"
                stack.append((child, depth + 1))
        elif isinstance(value, list):
            if len(value) + nodes + len(stack) > MAX_METADATA_NODES:
                return "input_too_large"
            stack.extend((child, depth + 1) for child in value)
        elif isinstance(value, str):
            if len(value) > MAX_METADATA_BYTES:
                return "input_too_large"
            if _utf8_size(value) is None:
                return "invalid_input"
        elif value is not None and not isinstance(value, (bool, int, float)):
            return "invalid_input"
    try:
        encoded = json.dumps(
            metadata,
            ensure_ascii=False,
            allow_nan=False,
            separators=(",", ":"),
        ).encode("utf-8")
    except (RecursionError, TypeError, UnicodeEncodeError, ValueError):
        return "invalid_input"
    return "input_too_large" if len(encoded) > MAX_METADATA_BYTES else None
 def _validate_payload(
    target: Any,
    content: Any,
    account_id: Any,
    message_id: Any,
    metadata: Any,
    dry_run: Any,
 ) -> Optional[str]:
    if not isinstance(target, str) or not isinstance(content, str):
        return "invalid_input"
    if not target.strip() or not content.strip() or not isinstance(dry_run, bool):
        return "invalid_input"
    if any(ord(character) < 32 or ord(character) == 127 for character in target):
        return "invalid_input"
    if "\x00" in content or not _valid_identifier(account_id) or not _valid_identifier(message_id):
        return "invalid_input"
    if len(target) > MAX_TARGET_BYTES or len(content) > MAX_CONTENT_BYTES:
        return "input_too_large"
    target_size = _utf8_size(target)
    content_size = _utf8_size(content)
    if target_size is None or content_size is None:
        return "invalid_input"
    if target_size > MAX_TARGET_BYTES or content_size > MAX_CONTENT_BYTES:
        return "input_too_large"
    return _validate_metadata(metadata)
 def _safe_provider_message_id(value: Any) -> Optional[str]:
    if not isinstance(value, str) or not value:
        return None
    if len(value) > MAX_IDENTIFIER_BYTES:
        return None
    size = _utf8_size(value)
    if size is None or size > MAX_IDENTIFIER_BYTES:
        return None
    if any(ord(character) < 32 or ord(character) == 127 for character in value):
        return None
    return value
 def send_ucc_adapter(
    target: str,
    content: str,
    account_id: str = "",
    message_id: str = "",
    metadata: Optional[Dict[str, Any]] = None,
    dry_run: bool = False,
 ) -> BackendResult:
    effective_account = account_id or WECHAT_ACCOUNT_ID
    effective_metadata: Any = {} if metadata is None else metadata
    error_code = _validate_payload(
        target,
        content,
        effective_account,
        message_id,
        effective_metadata,
        dry_run,
    )
    if error_code:
        return BackendResult(False, "ucc-adapter", "failed", {"code": error_code})
    payload = {
        "target": target,
        "content": content,
        "account_id": effective_account,
        "message_id": message_id,
        "metadata": effective_metadata,
        "dry_run": dry_run,
    }
    last_endpoint: Optional[str] = None
    for path in _ucc_send_paths():
        last_endpoint = path
        try:
            http_status, data = _post_ucc(path, payload)
        except urllib.error.HTTPError as exc:
            if exc.code in {404, 405}:
                continue
            return BackendResult(
                False,
                "ucc-adapter",
                "failed",
                {"code": "ucc_adapter_http_error", "http_status": int(exc.code), "endpoint": path},
            )
        except (OSError, ValueError, urllib.error.URLError, _BackendProtocolError):
            return BackendResult(
                False,
                "ucc-adapter",
                "failed",
                {"code": "ucc_adapter_unavailable", "endpoint": path},
            )
        except Exception:
            # Fail closed without returning exception text, type, or chained cause.
            return BackendResult(
                False,
                "ucc-adapter",
                "failed",
                {"code": "ucc_adapter_unavailable", "endpoint": path},
            )
        if not (200 <= http_status < 300):
            return BackendResult(
                False,
                "ucc-adapter",
                "failed",
                {"code": "ucc_adapter_http_error", "http_status": http_status, "endpoint": path},
            )
        if not _ucc_result_is_ok(data, dry_run):
            return BackendResult(
                False,
                "ucc-adapter",
                "failed",
                {"code": "ucc_adapter_rejected", "http_status": http_status, "endpoint": path},
            )
        confirmed_status = "dry_run" if dry_run else "sent"
        result_data: Dict[str, Any] = {
            "code": confirmed_status,
            "http_status": http_status,
            "endpoint": path,
        }
        provider_message_id = _safe_provider_message_id(data.get("provider_message_id"))
        if provider_message_id:
            result_data["provider_message_id"] = provider_message_id
        return BackendResult(True, "ucc-adapter", confirmed_status, result_data)
    data: Dict[str, Any] = {"code": "ucc_adapter_unavailable"}
    if last_endpoint:
        data["endpoint"] = last_endpoint
    return BackendResult(False, "ucc-adapter", "failed", data)
 def send_openclaw_channel(
    target: str,
    content: str,
    account_id: str = "",
    message_id: str = "",
    metadata: Optional[Dict[str, Any]] = None,
    dry_run: bool = False,
 ) -> BackendResult:
    del target, content, account_id, message_id, metadata, dry_run
    return BackendResult(
        False,
        "openclaw-channel",
        "failed",
        {"code": "unsupported_backend"},
    )
 def send_message(
    target: str,
    content: str,
    account_id: str = "",
    message_id: str = "",
    metadata: Optional[Dict[str, Any]] = None,
    dry_run: bool = False,
 ) -> Dict[str, Any]:
    selected = select_backend()
    if selected.get("backend") == "ucc-adapter":
        return send_ucc_adapter(
            target, content, account_id, message_id, metadata, dry_run
        ).to_dict()
    code = "unsupported_backend" if selected.get("reason") == "unsupported_backend" else "wechat_outbound_backend_unavailable"
    return BackendResult(False, "unavailable", "failed", {"code": code}).to_dict()
 if __name__ == "__main__":
    print(json.dumps(select_backend(), ensure_ascii=False, separators=(",", ":")))
@@ -1 +0,0 @@
 """AIOS services namespace."""
@@ -1,5 +0,0 @@
 """Reliable, fail-closed Codex command watcher."""
 from .watcher import CommandWatcher, Config, load_config
 __all__ = ["CommandWatcher", "Config", "load_config"]
@@ -1,19 +0,0 @@
 from __future__ import annotations
 import argparse
 import sys
 from .watcher import CommandWatcher, load_config
 def main() -> int:
    parser = argparse.ArgumentParser(description="Claim and execute Codex work units")
    parser.add_argument("--config", required=True)
    parser.add_argument("--once", action="store_true", help="process one scan and exit")
    args = parser.parse_args()
    watcher = CommandWatcher(load_config(args.config))
    return watcher.run_once() if args.once else watcher.run_forever()
 if __name__ == "__main__":
    sys.exit(main())
@@ -1,480 +0,0 @@
 from __future__ import annotations
 from dataclasses import dataclass
 import hashlib
 import json
 import os
 from pathlib import Path
 import re
 import sqlite3
 import stat
 import subprocess
 import time
 from typing import Any, Callable
 import urllib.error
 import urllib.parse
 import urllib.request
 import uuid
 OID_RE = re.compile(r"^[0-9a-f]{40}$")
 ID_RE = re.compile(r"^[A-Za-z0-9][A-Za-z0-9._:-]{0,127}$")
 TERMINAL = {"COMPLETED", "DEAD_LETTER", "REJECTED"}
 class WatcherError(RuntimeError):
    pass
@dataclass(frozen=True)
 class Config:
    active_dir: Path
    completed_dir: Path
    dead_letter_dir: Path
    database_path: Path
    audit_log_path: Path
    gateway_endpoint: str
    gateway_token_file: Path
    contract_commit: str
    lease_seconds: int
    retry_seconds: int
    max_attempts: int
    poll_seconds: float
    executor_path: Path | None
    executor_args: tuple[str, ...]
    executor_timeout_seconds: int
 def _absolute(value: Any, key: str) -> Path:
    if not isinstance(value, str) or not value or not Path(value).is_absolute():
        raise WatcherError(f"invalid_{key}")
    return Path(value)
 def _secure_regular(path: Path, *, writable_mask: int = 0o022) -> os.stat_result:
    info = os.lstat(path)
    if stat.S_ISLNK(info.st_mode) or not stat.S_ISREG(info.st_mode):
        raise WatcherError(f"unsafe_file:{path}")
    if stat.S_IMODE(info.st_mode) & writable_mask:
        raise WatcherError(f"unsafe_mode:{path}")
    return info
 def _read_secure_text(
    path: Path,
    *,
    writable_mask: int = 0o022,
    max_bytes: int,
    allowed_owners: set[int] | None = None,
 ) -> str:
    flags = os.O_RDONLY | getattr(os, "O_NOFOLLOW", 0)
    fd = os.open(path, flags)
    try:
        info = os.fstat(fd)
        if not stat.S_ISREG(info.st_mode) or stat.S_IMODE(info.st_mode) & writable_mask:
            raise WatcherError(f"unsafe_file:{path}")
        if allowed_owners is not None and info.st_uid not in allowed_owners:
            raise WatcherError(f"owner_mismatch:{path}")
        if info.st_size > max_bytes:
            raise WatcherError(f"file_too_large:{path}")
        chunks: list[bytes] = []
        total = 0
        while True:
            chunk = os.read(fd, min(65_536, max_bytes + 1 - total))
            if not chunk:
                break
            chunks.append(chunk)
            total += len(chunk)
            if total > max_bytes:
                raise WatcherError(f"file_too_large:{path}")
        return b"".join(chunks).decode("utf-8")
    finally:
        os.close(fd)
 def load_config(path_value: str) -> Config:
    path = _absolute(path_value, "config_path")
    raw = json.loads(_read_secure_text(path, max_bytes=131_072, allowed_owners={0, os.geteuid()}))
    expected = {
        "active_dir", "completed_dir", "dead_letter_dir", "database_path",
        "audit_log_path", "gateway_endpoint", "gateway_token_file",
        "contract_commit", "lease_seconds", "retry_seconds", "max_attempts",
        "poll_seconds", "executor_path", "executor_args",
        "executor_timeout_seconds",
    }
    if not isinstance(raw, dict) or set(raw) != expected:
        raise WatcherError("invalid_config_fields")
    contract = raw["contract_commit"]
    if not isinstance(contract, str) or not OID_RE.fullmatch(contract):
        raise WatcherError("invalid_contract_commit")
    endpoint = raw["gateway_endpoint"]
    if not isinstance(endpoint, str) or not endpoint.startswith("http://127.0.0.1:") or not endpoint.endswith("/v1/orchestrator/events"):
        raise WatcherError("gateway_endpoint_must_be_loopback")
    executor_raw = raw["executor_path"]
    executor = None if executor_raw is None else _absolute(executor_raw, "executor_path")
    executor_args = raw["executor_args"]
    if not isinstance(executor_args, list) or any(not isinstance(x, str) or not x for x in executor_args):
        raise WatcherError("invalid_executor_args")
    ints = ("lease_seconds", "retry_seconds", "max_attempts", "executor_timeout_seconds")
    if any(not isinstance(raw[x], int) or raw[x] <= 0 for x in ints):
        raise WatcherError("invalid_positive_integer")
    if not isinstance(raw["poll_seconds"], (int, float)) or raw["poll_seconds"] <= 0:
        raise WatcherError("invalid_poll_seconds")
    return Config(
        active_dir=_absolute(raw["active_dir"], "active_dir"),
        completed_dir=_absolute(raw["completed_dir"], "completed_dir"),
        dead_letter_dir=_absolute(raw["dead_letter_dir"], "dead_letter_dir"),
        database_path=_absolute(raw["database_path"], "database_path"),
        audit_log_path=_absolute(raw["audit_log_path"], "audit_log_path"),
        gateway_endpoint=endpoint,
        gateway_token_file=_absolute(raw["gateway_token_file"], "gateway_token_file"),
        contract_commit=contract,
        lease_seconds=raw["lease_seconds"], retry_seconds=raw["retry_seconds"],
        max_attempts=raw["max_attempts"], poll_seconds=float(raw["poll_seconds"]),
        executor_path=executor, executor_args=tuple(executor_args),
        executor_timeout_seconds=raw["executor_timeout_seconds"],
    )
 def canonical_digest(value: dict[str, Any]) -> str:
    return hashlib.sha256(json.dumps(value, sort_keys=True, separators=(",", ":")).encode()).hexdigest()
 def validate_command(value: Any, expected_contract: str) -> dict[str, Any]:
    expected = {
        "schema_version", "command_id", "project_key", "work_unit_id",
        "contract_commit", "state_version", "execution",
    }
    if not isinstance(value, dict) or set(value) != expected or value.get("schema_version") != 1:
        raise WatcherError("invalid_command_fields")
    for key in ("command_id", "project_key", "work_unit_id"):
        if not isinstance(value[key], str) or not ID_RE.fullmatch(value[key]):
            raise WatcherError(f"invalid_{key}")
    if value["contract_commit"] != expected_contract:
        raise WatcherError("contract_commit_mismatch")
    if not isinstance(value["state_version"], int) or value["state_version"] < 0:
        raise WatcherError("invalid_state_version")
    execution = value["execution"]
    if not isinstance(execution, dict) or set(execution) != {"kind", "input"}:
        raise WatcherError("invalid_execution")
    if execution["kind"] not in {"acceptance_probe", "codex"}:
        raise WatcherError("invalid_execution_kind")
    if not isinstance(execution["input"], str) or len(execution["input"].encode()) > 65_536:
        raise WatcherError("invalid_execution_input")
    return value
 class CommandWatcher:
    def __init__(
        self,
        config: Config,
        *,
        clock: Callable[[], float] = time.time,
        event_poster: Callable[[dict[str, Any]], None] | None = None,
        executor: Callable[[dict[str, Any]], dict[str, Any]] | None = None,
        worker_id: str | None = None,
    ) -> None:
        self.config = config
        self.clock = clock
        self.worker_id = worker_id or f"{os.uname().nodename}:{os.getpid()}:{uuid.uuid4().hex[:8]}"
        self._uses_gateway = event_poster is None
        self.event_poster = event_poster or self._post_event
        self.executor = executor or self._execute
        self._prepare()
    def _prepare(self) -> None:
        for directory in (self.config.active_dir, self.config.completed_dir, self.config.dead_letter_dir):
            info = os.lstat(directory)
            if stat.S_ISLNK(info.st_mode) or not stat.S_ISDIR(info.st_mode) or stat.S_IMODE(info.st_mode) & 0o002:
                raise WatcherError(f"unsafe_directory:{directory}")
        self.config.database_path.parent.mkdir(parents=True, exist_ok=True, mode=0o700)
        self.config.audit_log_path.parent.mkdir(parents=True, exist_ok=True, mode=0o700)
        with self._db() as db:
            db.execute("PRAGMA journal_mode=WAL")
            db.execute("""
                CREATE TABLE IF NOT EXISTS commands (
                    command_id TEXT PRIMARY KEY, work_unit_id TEXT NOT NULL UNIQUE,
                    payload_sha256 TEXT NOT NULL,
                    source_path TEXT NOT NULL, status TEXT NOT NULL,
                    phase TEXT NOT NULL, attempts INTEGER NOT NULL DEFAULT 0,
                    lease_owner TEXT, lease_until REAL, next_attempt REAL NOT NULL DEFAULT 0,
                    result_json TEXT, start_event_json TEXT, end_event_json TEXT,
                    last_error TEXT, updated_at REAL NOT NULL
                )
            """)
    def _db(self) -> sqlite3.Connection:
        db = sqlite3.connect(self.config.database_path, timeout=5, isolation_level=None)
        db.row_factory = sqlite3.Row
        return db
    def _audit(self, marker: str, command_id: str, **fields: Any) -> None:
        safe = {k: v for k, v in fields.items() if k not in {"input", "token", "secret"}}
        record = {"time": int(self.clock()), "marker": marker, "command_id": command_id, **safe}
        fd = os.open(self.config.audit_log_path, os.O_APPEND | os.O_CREAT | os.O_WRONLY | os.O_NOFOLLOW, 0o600)
        try:
            os.write(fd, (json.dumps(record, sort_keys=True, separators=(",", ":")) + "\n").encode())
        finally:
            os.close(fd)
    def _read_command(self, path: Path) -> tuple[dict[str, Any], str]:
        raw = _read_secure_text(path, max_bytes=131_072, allowed_owners={0, os.geteuid()})
        value = validate_command(json.loads(raw), self.config.contract_commit)
        return value, canonical_digest(value)
    def _register(self, command: dict[str, Any], digest: str, path: Path) -> str:
        now = self.clock()
        with self._db() as db:
            db.execute("BEGIN IMMEDIATE")
            row = db.execute("SELECT payload_sha256,status FROM commands WHERE command_id=?", (command["command_id"],)).fetchone()
            if row:
                db.execute("COMMIT")
                if row["payload_sha256"] != digest:
                    raise WatcherError("duplicate_command_payload_conflict")
                return "duplicate_terminal" if row["status"] in TERMINAL else "existing"
            prior_work_unit = db.execute(
                "SELECT command_id FROM commands WHERE work_unit_id=?",
                (command["work_unit_id"],),
            ).fetchone()
            if prior_work_unit:
                db.execute("COMMIT")
                raise WatcherError("duplicate_work_unit")
            db.execute(
                "INSERT INTO commands(command_id,work_unit_id,payload_sha256,source_path,status,phase,updated_at) VALUES(?,?,?,?,?,?,?)",
                (command["command_id"], command["work_unit_id"], digest, str(path), "PENDING", "START", now),
            )
            db.execute("COMMIT")
        return "new"
    def _claim(self, command_id: str) -> sqlite3.Row | None:
        now = self.clock()
        with self._db() as db:
            db.execute("BEGIN IMMEDIATE")
            row = db.execute("SELECT * FROM commands WHERE command_id=?", (command_id,)).fetchone()
            if not row or row["status"] in TERMINAL or row["next_attempt"] > now:
                db.execute("COMMIT")
                return None
            if row["lease_until"] is not None and row["lease_until"] > now and row["lease_owner"] != self.worker_id:
                db.execute("COMMIT")
                return None
            attempts = row["attempts"] + 1
            db.execute(
                "UPDATE commands SET status='CLAIMED',attempts=?,lease_owner=?,lease_until=?,updated_at=? WHERE command_id=?",
                (attempts, self.worker_id, now + self.config.lease_seconds, now, command_id),
            )
            row = db.execute("SELECT * FROM commands WHERE command_id=?", (command_id,)).fetchone()
            db.execute("COMMIT")
        assert row is not None
        self._audit("CODEX_AUTO_CLAIM", command_id, attempt=row["attempts"], worker=self.worker_id)
        return row
    def _update(self, command_id: str, *, status: str, phase: str, result: dict[str, Any] | None = None, error: str | None = None) -> None:
        with self._db() as db:
            release = status in TERMINAL
            db.execute(
                "UPDATE commands SET status=?,phase=?,result_json=COALESCE(?,result_json),last_error=?,lease_owner=CASE WHEN ? THEN NULL ELSE lease_owner END,lease_until=CASE WHEN ? THEN NULL ELSE lease_until END,updated_at=? WHERE command_id=?",
                (status, phase, json.dumps(result, sort_keys=True) if result is not None else None, error, release, release, self.clock(), command_id),
            )
    def _retry_or_dead(self, command_id: str, path: Path, error: Exception) -> None:
        message = f"{type(error).__name__}:{str(error)[:240]}"
        with self._db() as db:
            row = db.execute("SELECT attempts,phase FROM commands WHERE command_id=?", (command_id,)).fetchone()
            attempts = row["attempts"] if row else self.config.max_attempts
            if str(error) == "gateway_stale_state_version" and row:
                column = "start_event_json" if row["phase"] == "START" else "end_event_json"
                db.execute(f"UPDATE commands SET {column}=NULL WHERE command_id=?", (command_id,))
            if attempts >= self.config.max_attempts:
                db.execute("UPDATE commands SET status='DEAD_LETTER',last_error=?,lease_owner=NULL,lease_until=NULL,updated_at=? WHERE command_id=?", (message, self.clock(), command_id))
                self._move(path, self.config.dead_letter_dir)
                self._audit("DEAD_LETTER", command_id, error=message)
            else:
                db.execute("UPDATE commands SET status='RETRY',last_error=?,next_attempt=?,lease_owner=NULL,lease_until=NULL,updated_at=? WHERE command_id=?", (message, self.clock() + self.config.retry_seconds, self.clock(), command_id))
                self._audit("RETRY_SCHEDULED", command_id, attempt=attempts, error=message)
    def _event(self, command: dict[str, Any], event_type: str, status: str, suffix: str, result: dict[str, Any] | None = None) -> dict[str, Any]:
        return {
            "schema_version": 1, "project_key": command["project_key"],
            "contract_commit": command["contract_commit"],
            "event_id": f"{command['command_id']}:{suffix}", "source_agent": "codex",
            "event_type": event_type, "work_unit_id": command["work_unit_id"],
            "status": status, "state_version": command["state_version"],
            "implementation_commit": None, "result_commit": None, "blocker": None,
            "evidence_refs": [], "payload": {"command_id": command["command_id"], "result": result or {}},
        }
    def _gateway_token(self) -> str:
        token = _read_secure_text(
            self.config.gateway_token_file,
            writable_mask=0o027,
            max_bytes=16_384,
            allowed_owners={0, os.geteuid()},
        )
        if not token or "\n" in token or "\r" in token:
            raise WatcherError("invalid_gateway_token")
        return token
    def _gateway_state_version(self, project_key: str, fallback: int) -> int:
        state_url = self.config.gateway_endpoint.removesuffix("/events") + "/state/" + urllib.parse.quote(project_key, safe="")
        request = urllib.request.Request(
            state_url,
            headers={"Authorization": f"Bearer {self._gateway_token()}"},
            method="GET",
        )
        try:
            with urllib.request.urlopen(request, timeout=15) as response:
                payload = json.loads(response.read(262_145))
                version = payload.get("state_version")
                if not isinstance(version, int) or version < 0:
                    raise WatcherError("invalid_gateway_state")
                return version
        except urllib.error.HTTPError as exc:
            if exc.code == 404:
                return fallback
            raise WatcherError(f"gateway_state_http_{exc.code}") from exc
        except urllib.error.URLError as exc:
            raise WatcherError("gateway_state_unavailable") from exc
    def _durable_event(self, command: dict[str, Any], column: str, event: dict[str, Any]) -> dict[str, Any]:
        with self._db() as db:
            row = db.execute(f"SELECT {column} FROM commands WHERE command_id=?", (command["command_id"],)).fetchone()
            if row and row[column]:
                return json.loads(row[column])
        if self._uses_gateway:
            event["state_version"] = self._gateway_state_version(command["project_key"], command["state_version"])
        encoded = json.dumps(event, sort_keys=True, separators=(",", ":"))
        with self._db() as db:
            db.execute(f"UPDATE commands SET {column}=? WHERE command_id=? AND {column} IS NULL", (encoded, command["command_id"]))
            saved = db.execute(f"SELECT {column} FROM commands WHERE command_id=?", (command["command_id"],)).fetchone()
        if not saved or not saved[column]:
            raise WatcherError("event_persistence_failed")
        return json.loads(saved[column])
    def _post_event(self, event: dict[str, Any]) -> None:
        request = urllib.request.Request(
            self.config.gateway_endpoint,
            data=json.dumps(event, separators=(",", ":")).encode(),
            headers={"Authorization": f"Bearer {self._gateway_token()}", "Content-Type": "application/json"},
            method="POST",
        )
        try:
            with urllib.request.urlopen(request, timeout=15) as response:
                if response.status not in {200, 201, 202}:
                    raise WatcherError(f"gateway_http_{response.status}")
        except urllib.error.HTTPError as exc:
            body = exc.read(16_384)
            try:
                reason = json.loads(body).get("error")
            except (json.JSONDecodeError, UnicodeDecodeError, AttributeError):
                reason = None
            if exc.code == 409 and reason == "stale_state_version":
                raise WatcherError("gateway_stale_state_version") from exc
            raise WatcherError(f"gateway_http_{exc.code}") from exc
        except urllib.error.URLError as exc:
            raise WatcherError("gateway_unavailable") from exc
    def _execute(self, command: dict[str, Any]) -> dict[str, Any]:
        execution = command["execution"]
        if execution["kind"] == "acceptance_probe":
            return {"kind": "acceptance_probe", "input_sha256": hashlib.sha256(execution["input"].encode()).hexdigest(), "exit_code": 0}
        assert self.config.executor_path is not None
        proc = subprocess.run(
            [str(self.config.executor_path), *self.config.executor_args],
            input=execution["input"], text=True, stdout=subprocess.DEVNULL,
            stderr=subprocess.DEVNULL, timeout=self.config.executor_timeout_seconds,
            check=False, env={"PATH": "/usr/bin:/bin", "LANG": "C.UTF-8"},
        )
        if proc.returncode != 0:
            raise WatcherError(f"codex_executor_failed:{proc.returncode}")
        return {"kind": "codex", "exit_code": 0}
    def _preflight_execution(self, command: dict[str, Any]) -> None:
        if command["execution"]["kind"] == "acceptance_probe":
            return
        if self.config.executor_path is None:
            raise WatcherError("codex_executor_unavailable")
        info = _secure_regular(self.config.executor_path)
        if not info.st_mode & stat.S_IXUSR:
            raise WatcherError("codex_executor_not_executable")
    def _move(self, source: Path, target_dir: Path) -> None:
        if not source.exists() and not source.is_symlink():
            return
        target = target_dir / source.name
        if target.exists() or target.is_symlink():
            target = target_dir / f"{source.stem}-{uuid.uuid4().hex[:8]}{source.suffix}"
        os.replace(source, target)
    def _reject_unregistered(self, path: Path, error: Exception) -> None:
        command_id = path.stem[:128] if ID_RE.fullmatch(path.stem[:128]) else f"invalid-{uuid.uuid4().hex[:12]}"
        self._audit("COMMAND_REJECTED", command_id, error=f"{type(error).__name__}:{str(error)[:240]}")
        self._move(path, self.config.dead_letter_dir)
    def process_path(self, path: Path) -> bool:
        try:
            command, digest = self._read_command(path)
            registration = self._register(command, digest, path)
            if registration == "duplicate_terminal":
                self._move(path, self.config.completed_dir)
                return False
            row = self._claim(command["command_id"])
            if row is None:
                return False
            phase = row["phase"]
            result = json.loads(row["result_json"]) if row["result_json"] else None
            if phase == "START":
                start_event = self._durable_event(
                    command,
                    "start_event_json",
                    self._event(command, "work_unit_started", "RUNNING", "started"),
                )
                self.event_poster(start_event)
                self._audit("TASK_STARTED", command["command_id"])
                self._update(command["command_id"], status="PENDING", phase="EXECUTE")
                phase = "EXECUTE"
            if phase == "EXECUTE":
                self._preflight_execution(command)
                self._update(command["command_id"], status="CLAIMED", phase="EXECUTING")
                phase = "EXECUTING"
                result = self.executor(command)
                self._audit("TASK_EXECUTED", command["command_id"], result=result)
                self._update(command["command_id"], status="PENDING", phase="END", result=result)
                phase = "END"
            elif phase == "EXECUTING":
                raise WatcherError("ambiguous_execution_not_repeated")
            if phase == "END":
                assert result is not None
                end_event = self._durable_event(
                    command,
                    "end_event_json",
                    self._event(command, "work_unit_ended", "PASS", "ended", result),
                )
                self.event_poster(end_event)
                self._audit("TASK_ENDED", command["command_id"])
                self._audit("EVENT_POSTED", command["command_id"], event="work_unit_ended")
                self._update(command["command_id"], status="COMPLETED", phase="DONE", result=result)
                self._move(path, self.config.completed_dir)
            return True
        except Exception as exc:
            try:
                if isinstance(exc, WatcherError) and str(exc) in {"duplicate_command_payload_conflict", "duplicate_work_unit"}:
                    self._reject_unregistered(path, exc)
                elif "command" in locals():
                    self._retry_or_dead(command["command_id"], path, exc)
                else:
                    self._reject_unregistered(path, exc)
            except Exception:
                pass
            return False
    def run_once(self) -> int:
        failures = 0
        for path in sorted(self.config.active_dir.glob("*.json")):
            if not self.process_path(path):
                failures += 1
        return 1 if failures else 0
    def run_forever(self) -> int:
        while True:
            self.run_once()
            time.sleep(self.config.poll_seconds)
@@ -1,3 +0,0 @@
 """Reliable central orchestration gateway for AIOS."""
 __version__ = "0.2.0"
@@ -1,34 +0,0 @@
 from __future__ import annotations
 import argparse
 import logging
 from pathlib import Path
 from .config import Config
 from .service import Runtime, serve
 def main() -> int:
    parser = argparse.ArgumentParser(description="AIOS Orchestrator Gateway")
    parser.add_argument("--config", type=Path, required=True)
    parser.add_argument("--log-level", default="INFO")
    parser.add_argument(
        "--poll-once",
        action="store_true",
        help="Run one inbox/model/outbox/OpenClaw synchronization polling cycle.",
    )
    args = parser.parse_args()
    logging.basicConfig(
        level=getattr(logging, args.log_level.upper(), logging.INFO),
        format="%(asctime)s %(levelname)s %(name)s %(message)s",
    )
    if args.poll_once:
        runtime = Runtime(Config.from_file(args.config))
        runtime.poll_once()
        return 0
    serve(args.config)
    return 0
 if __name__ == "__main__":
    raise SystemExit(main())
@@ -1,132 +0,0 @@
 from __future__ import annotations
 import json
 import os
 from pathlib import Path
 import re
 import stat
 import subprocess
 from typing import Any, Callable
 import urllib.error
 import urllib.request
 from .models import Decision
 GATEWAY_DECISION_URL = "http://127.0.0.1:18920/v1/orchestrator/decisions"
 SECRET_REF_RE = re.compile(r"^secret://[A-Za-z0-9._/-]{1,480}$")
 class CallbackBridgeError(RuntimeError):
    def __init__(self, category: str, status_code: int) -> None:
        super().__init__(category)
        self.category = category
        self.status_code = status_code
 def _trusted_resolver(path: Path) -> Path:
    if not path.is_absolute():
        raise CallbackBridgeError("CALLBACK_CONFIGURATION_UNAVAILABLE", 503)
    try:
        info = path.lstat()
        resolved = path.resolve(strict=True)
    except OSError as exc:
        raise CallbackBridgeError("CALLBACK_CONFIGURATION_UNAVAILABLE", 503) from exc
    if (
        resolved != path
        or stat.S_ISLNK(info.st_mode)
        or not stat.S_ISREG(info.st_mode)
        or info.st_uid != 0
        or stat.S_IMODE(info.st_mode) & 0o022
        or not stat.S_IMODE(info.st_mode) & 0o111
    ):
        raise CallbackBridgeError("CALLBACK_CONFIGURATION_UNAVAILABLE", 503)
    return resolved
 def resolve_secret_ref(secret_ref: str, resolver: Path) -> str:
    if not isinstance(secret_ref, str) or not SECRET_REF_RE.fullmatch(secret_ref):
        raise CallbackBridgeError("CALLBACK_CONFIGURATION_UNAVAILABLE", 503)
    executable = _trusted_resolver(resolver)
    request = json.dumps(
        {"operation": "resolve", "secret_ref": secret_ref},
        separators=(",", ":"),
        sort_keys=True,
    ).encode()
    try:
        result = subprocess.run(
            [str(executable)],
            input=request,
            stdout=subprocess.PIPE,
            stderr=subprocess.DEVNULL,
            env={"PATH": "/usr/bin:/bin", "LANG": "C", "LC_ALL": "C"},
            timeout=5,
            check=False,
        )
    except (OSError, subprocess.SubprocessError) as exc:
        raise CallbackBridgeError("CALLBACK_CREDENTIAL_UNAVAILABLE", 503) from exc
    token = result.stdout.removesuffix(b"\n")
    if (
        result.returncode != 0
        or not token
        or len(token) > 16_384
        or b"\n" in token
        or b"\r" in token
    ):
        raise CallbackBridgeError("CALLBACK_CREDENTIAL_UNAVAILABLE", 503)
    try:
        return token.decode("utf-8")
    except UnicodeDecodeError as exc:
        raise CallbackBridgeError("CALLBACK_CREDENTIAL_UNAVAILABLE", 503) from exc
 def submit_orchestration_decision(
    payload: dict[str, Any],
    *,
    contract_commit: str,
    secret_ref: str,
    resolver: Path,
    request_open: Callable[..., Any] = urllib.request.urlopen,
 ) -> dict[str, Any]:
    try:
        decision = Decision.from_dict(payload)
    except Exception as exc:
        raise CallbackBridgeError("INVALID_DECISION", 422) from exc
    if decision.contract_commit != contract_commit:
        raise CallbackBridgeError("CONTRACT_MISMATCH", 409)
    token = resolve_secret_ref(secret_ref, resolver)
    request = urllib.request.Request(
        GATEWAY_DECISION_URL,
        data=json.dumps(
            decision.to_dict(), separators=(",", ":"), sort_keys=True
        ).encode(),
        method="POST",
        headers={
            "Authorization": f"Bearer {token}",
            "Content-Type": "application/json",
            "Idempotency-Key": decision.event_id,
            "User-Agent": "aios-chatgpt-actions/decision-bridge-v1",
        },
    )
    try:
        with request_open(request, timeout=10) as response:
            raw = response.read(65_537)
            if response.status != 202 or len(raw) > 65_536:
                raise CallbackBridgeError("CALLBACK_HTTP_ERROR", 502)
    except urllib.error.HTTPError as exc:
        if exc.code == 409:
            raise CallbackBridgeError("STALE_STATE", 409) from exc
        raise CallbackBridgeError("CALLBACK_HTTP_ERROR", 502) from exc
    except (urllib.error.URLError, TimeoutError, OSError) as exc:
        raise CallbackBridgeError("CALLBACK_NETWORK_ERROR", 502) from exc
    try:
        value = json.loads(raw)
    except json.JSONDecodeError as exc:
        raise CallbackBridgeError("CALLBACK_RESPONSE_INVALID", 502) from exc
    if (
        not isinstance(value, dict)
        or value.get("accepted") is not True
        or not isinstance(value.get("state_version"), int)
    ):
        raise CallbackBridgeError("CALLBACK_RESPONSE_INVALID", 502)
    return {"accepted": True, "state_version": value["state_version"]}
@@ -1,117 +0,0 @@
 from __future__ import annotations
 from dataclasses import dataclass
 import ipaddress
 from typing import Any
 from urllib.parse import urlparse
 from .errors import ConfigurationError
 from .models import DECISION_SCHEMA
 TRANSPORT_MODES = {"remote_https_action", "secure_mcp_tunnel"}
 FORBIDDEN_HOSTS = {"localhost", "localhost.localdomain"}
 FORBIDDEN_MARKERS = {"replace_with_actions_host", "placeholder"}
 def _https_endpoint(value: Any, *, public: bool) -> str:
    if not isinstance(value, str) or not value or any(
        marker in value.lower() for marker in FORBIDDEN_MARKERS
    ):
        raise ConfigurationError("invalid_callback_endpoint")
    parsed = urlparse(value)
    host = (parsed.hostname or "").lower()
    if (
        parsed.scheme != "https"
        or not host
        or parsed.username is not None
        or parsed.password is not None
        or parsed.query
        or parsed.fragment
        or host in FORBIDDEN_HOSTS
    ):
        raise ConfigurationError("invalid_callback_endpoint")
    try:
        address = ipaddress.ip_address(host)
    except ValueError:
        address = None
    if address is not None and (
        address.is_loopback
        or address.is_private
        or address.is_unspecified
        or address.is_link_local
        or address.is_multicast
        or address.is_reserved
    ):
        raise ConfigurationError("invalid_callback_endpoint")
    if public and (address is None and ("." not in host or host.endswith((".local", ".internal")))):
        raise ConfigurationError("callback_endpoint_not_public")
    return value.rstrip("/")
@dataclass(frozen=True)
 class CallbackTransport:
    mode: str
    endpoint: str
    approved_mcp_tunnel_endpoints: tuple[str, ...]
    @classmethod
    def from_dict(cls, value: dict[str, Any]) -> "CallbackTransport":
        expected = {"mode", "endpoint", "approved_mcp_tunnel_endpoints"}
        if not isinstance(value, dict) or set(value) != expected:
            raise ConfigurationError("invalid_callback_transport_fields")
        mode = value.get("mode")
        if mode not in TRANSPORT_MODES:
            raise ConfigurationError("invalid_callback_transport_mode")
        approved = value.get("approved_mcp_tunnel_endpoints")
        if not isinstance(approved, list) or any(not isinstance(x, str) for x in approved):
            raise ConfigurationError("invalid_approved_mcp_tunnel_endpoints")
        normalized_approved = tuple(_https_endpoint(item, public=False) for item in approved)
        endpoint = _https_endpoint(value.get("endpoint"), public=mode == "remote_https_action")
        if mode == "secure_mcp_tunnel" and endpoint not in normalized_approved:
            raise ConfigurationError("unapproved_secure_mcp_tunnel_endpoint")
        return cls(mode, endpoint, normalized_approved)
    def action_openapi(self) -> dict[str, Any]:
        if self.mode != "remote_https_action":
            raise ConfigurationError("action_manifest_for_wrong_transport")
        return {
            "openapi": "3.1.0",
            "info": {"title": "AIOS Orchestrator Decision Callback", "version": "1.1.0"},
            "servers": [{"url": self.endpoint}],
            "paths": {
                "/v1/orchestrator/decisions": {
                    "post": {
                        "operationId": "submit_orchestration_decision",
                        "summary": "Submit one strict orchestration decision",
                        "security": [{"bearerAuth": []}],
                        "requestBody": {
                            "required": True,
                            "content": {"application/json": {"schema": DECISION_SCHEMA}},
                        },
                        "responses": {"202": {"description": "Accepted"}},
                    }
                }
            },
            "components": {
                "securitySchemes": {
                    "bearerAuth": {"type": "http", "scheme": "bearer"}
                }
            },
        }
    def mcp_tool(self) -> dict[str, Any]:
        if self.mode != "secure_mcp_tunnel":
            raise ConfigurationError("mcp_manifest_for_wrong_transport")
        return {
            "name": "submit_orchestration_decision",
            "description": "Validate and submit one AIOS orchestration decision.",
            "inputSchema": DECISION_SCHEMA,
            "annotations": {
                "title": "Submit orchestration decision",
                "readOnlyHint": False,
                "destructiveHint": False,
                "idempotentHint": True,
                "openWorldHint": False,
            },
        }
@@ -1,204 +0,0 @@
 from __future__ import annotations
 import json
 from dataclasses import dataclass
 import urllib.error
 import urllib.request
 from typing import Any
 from .errors import TriggerError
 from .models import DECISION_SCHEMA, Decision, Event
@dataclass(frozen=True)
 class WorkspaceTriggerResult:
    http_status: int
    remote_id: None = None
 WORKSPACE_HTTP_CATEGORIES = {
    401: "UNAUTHORIZED",
    403: "FORBIDDEN",
    404: "NOT_FOUND",
    409: "NOT_RUNNABLE",
 }
 class WorkspaceAgentClient:
    """Asynchronous Workspace Agent trigger.
    The official endpoint returns only 202 and no run ID or retrievable output.
    Completion therefore arrives through the gateway callback endpoint.
    """
    def __init__(self, trigger_id: str, token: str, timeout_seconds: int = 30):
        self.trigger_id = trigger_id
        self.token = token
        self.timeout_seconds = timeout_seconds
    def trigger(self, event: Event) -> WorkspaceTriggerResult:
        url = f"https://api.chatgpt.com/v1/workspace_agents/{self.trigger_id}/trigger"
        prompt = (
            "Process this AIOS orchestration event under the committed contract. "
            "When complete, call submit_orchestration_decision with strict JSON. "
            "Do not fabricate tests, Git state, secrets, evidence, or deployment facts.\n\n"
            + event.canonical_json()
        )
        body = json.dumps(
            {
                "conversation_key": event.project_key,
                "input": prompt,
            }
        ).encode()
        request = urllib.request.Request(
            url,
            data=body,
            method="POST",
            headers={
                "Authorization": f"Bearer {self.token}",
                "Content-Type": "application/json",
                "Idempotency-Key": event.event_id,
                "User-Agent": "aios-orchestrator-gateway/0.2",
            },
        )
        try:
            with urllib.request.urlopen(request, timeout=self.timeout_seconds) as response:
                if response.status != 202:
                    raise TriggerError(
                        "workspace_agent_unexpected_http_status",
                        http_status=response.status,
                    )
                return WorkspaceTriggerResult(http_status=202)
        except urllib.error.HTTPError as exc:
            category = WORKSPACE_HTTP_CATEGORIES.get(exc.code)
            raise TriggerError(
                "workspace_agent_http_error",
                http_status=exc.code,
                category=category,
            ) from exc
        except (urllib.error.URLError, TimeoutError, OSError) as exc:
            raise TriggerError(
                "workspace_agent_network_error",
                category="NETWORK_ERROR",
            ) from exc
 class ResponsesClient:
    """Background Responses API client with polling and strict output validation."""
    def __init__(self, api_key: str, model: str, timeout_seconds: int = 60):
        self.api_key = api_key
        self.model = model
        self.timeout_seconds = timeout_seconds
    def start(self, event: Event) -> str:
        body = {
            "model": self.model,
            "background": True,
            "instructions": (
                "You are the AIOS central orchestration brain. Apply the exact "
                "contract_commit in the event. Return only a strict decision. "
                "Never fabricate evidence, secrets, Git facts, tests, approvals, "
                "or deployment state. Keep external blockers fail-closed."
            ),
            "input": event.canonical_json(),
            "text": {
                "format": {
                    "type": "json_schema",
                    "name": "aios_orchestrator_decision",
                    "strict": True,
                    "schema": DECISION_SCHEMA,
                }
            },
            "metadata": {
                "event_id": event.event_id,
                "project_key": event.project_key,
                "contract_commit": event.contract_commit,
            },
        }
        payload = self._request(
            "POST",
            "https://api.openai.com/v1/responses",
            body,
            idempotency_key=event.event_id,
        )
        response_id = payload.get("id")
        if not isinstance(response_id, str) or not response_id.startswith("resp_"):
            raise TriggerError("responses_missing_id")
        return response_id
    def poll(self, response_id: str) -> tuple[str, Decision | None]:
        payload = self._request(
            "GET",
            f"https://api.openai.com/v1/responses/{response_id}",
            None,
        )
        status = payload.get("status")
        if status in {"queued", "in_progress"}:
            return str(status), None
        if status == "completed":
            text = _extract_output_text(payload)
            try:
                return "completed", Decision.from_dict(json.loads(text))
            except json.JSONDecodeError as exc:
                raise TriggerError("responses_invalid_json") from exc
        if status in {"failed", "cancelled", "incomplete"}:
            raise TriggerError(f"responses_{status}")
        raise TriggerError("responses_unknown_status")
    def _request(
        self,
        method: str,
        url: str,
        body: dict[str, Any] | None,
        *,
        idempotency_key: str | None = None,
    ) -> dict[str, Any]:
        encoded = json.dumps(body).encode() if body is not None else None
        headers = {
            "Authorization": f"Bearer {self.api_key}",
            "Content-Type": "application/json",
            "User-Agent": "aios-orchestrator-gateway/0.2",
        }
        if idempotency_key:
            headers["Idempotency-Key"] = idempotency_key
        request = urllib.request.Request(
            url,
            data=encoded,
            method=method,
            headers=headers,
        )
        try:
            with urllib.request.urlopen(request, timeout=self.timeout_seconds) as response:
                raw = response.read(4_194_304)
                if response.status not in {200, 201}:
                    raise TriggerError(f"responses_http_{response.status}")
        except urllib.error.HTTPError as exc:
            raise TriggerError(f"responses_http_{exc.code}") from exc
        except (urllib.error.URLError, TimeoutError, OSError) as exc:
            raise TriggerError("responses_request_failed") from exc
        try:
            payload = json.loads(raw)
        except json.JSONDecodeError as exc:
            raise TriggerError("responses_invalid_payload") from exc
        if not isinstance(payload, dict):
            raise TriggerError("responses_payload_not_object")
        return payload
 def _extract_output_text(payload: dict[str, Any]) -> str:
    output = payload.get("output")
    if not isinstance(output, list):
        raise TriggerError("responses_missing_output")
    for item in output:
        if not isinstance(item, dict):
            continue
        content = item.get("content")
        if not isinstance(content, list):
            continue
        for part in content:
            if isinstance(part, dict) and part.get("type") == "output_text":
                text = part.get("text")
                if isinstance(text, str):
                    return text
    raise TriggerError("responses_missing_output_text")
@@ -1,213 +0,0 @@
 from __future__ import annotations
 from dataclasses import dataclass
 import os
 from pathlib import Path
 import stat
 from typing import Any
 from .errors import ConfigurationError
 from .securefs import read_secure_bytes, read_secure_json, verify_secure_file
@dataclass(frozen=True)
 class Config:
    host: str
    port: int
    project_key: str
    database_path: Path
    event_inbox_dir: Path
    decision_inbox_dir: Path
    dispatch_outbox_dir: Path
    openclaw_sync_dir: Path
    command_queue_dir: Path
    inbound_token_file: Path
    callback_token_file: Path
    contract_commit: str
    workspace_trigger_id: str | None
    workspace_token_file: Path | None
    responses_api_key_file: Path | None
    responses_model: str | None
    poll_interval_seconds: int
    summary_interval_seconds: int
    callback_timeout_seconds: int
    lease_seconds: int
    retry_base_seconds: int
    max_retries: int
    max_same_blocker_retries: int
    max_hops_per_gate: int
    openclaw_sync_endpoint: str | None
    openclaw_sync_token_file: Path | None
    git_poll_enabled: bool
    git_poll_bare_repo: Path | None
    git_poll_ref_prefixes: tuple[str, ...]
    git_poll_interval_seconds: int
    @classmethod
    def from_file(cls, path: Path) -> "Config":
        owner_uid = int(os.environ.get("AIOS_ORCHESTRATOR_CONFIG_OWNER_UID", "0"))
        raw = read_secure_json(path, expected_uid=owner_uid)
        expected = {
            "host", "port", "project_key", "database_path", "event_inbox_dir",
            "decision_inbox_dir", "dispatch_outbox_dir", "openclaw_sync_dir",
            "command_queue_dir", "inbound_token_file", "callback_token_file",
            "contract_commit", "workspace_agent", "responses_fallback",
            "polling", "loop_controls", "openclaw_sync", "git_poll",
        }
        if set(raw) != expected:
            raise ConfigurationError("invalid_config_fields")
        def path_value(key: str) -> Path:
            value = raw.get(key)
            if not isinstance(value, str):
                raise ConfigurationError(f"invalid_{key}")
            result = Path(value)
            if not result.is_absolute():
                raise ConfigurationError(f"path_not_absolute:{key}")
            return result
        def integer(container: dict[str, Any], key: str, lo: int, hi: int) -> int:
            value = container.get(key)
            if not isinstance(value, int) or not lo <= value <= hi:
                raise ConfigurationError(f"invalid_{key}")
            return value
        host = raw.get("host")
        port = raw.get("port")
        project_key = raw.get("project_key")
        contract = raw.get("contract_commit")
        if not isinstance(host, str) or not host:
            raise ConfigurationError("invalid_host")
        if not isinstance(project_key, str) or not project_key:
            raise ConfigurationError("invalid_project_key")
        if not isinstance(port, int) or not 1 <= port <= 65535:
            raise ConfigurationError("invalid_port")
        if not isinstance(contract, str) or len(contract) != 40:
            raise ConfigurationError("invalid_contract_commit")
        workspace = raw.get("workspace_agent")
        responses = raw.get("responses_fallback")
        polling = raw.get("polling")
        loop = raw.get("loop_controls")
        sync = raw.get("openclaw_sync")
        git_poll = raw.get("git_poll")
        if not all(isinstance(x, dict) for x in (workspace, responses, polling, loop, sync, git_poll)):
            raise ConfigurationError("invalid_nested_config")
        trigger_id = workspace.get("trigger_id")
        workspace_token = workspace.get("access_token_file")
        if trigger_id is not None:
            if not isinstance(trigger_id, str) or not trigger_id.startswith("agtch_"):
                raise ConfigurationError("invalid_workspace_trigger_id")
            if not isinstance(workspace_token, str) or not Path(workspace_token).is_absolute():
                raise ConfigurationError("invalid_workspace_token_file")
            workspace_token_file = Path(workspace_token)
        else:
            workspace_token_file = None
        responses_key = responses.get("api_key_file")
        responses_model = responses.get("model")
        if responses_key is not None:
            if not isinstance(responses_key, str) or not Path(responses_key).is_absolute():
                raise ConfigurationError("invalid_responses_key_file")
            if not isinstance(responses_model, str) or not responses_model:
                raise ConfigurationError("invalid_responses_model")
            responses_key_file = Path(responses_key)
        else:
            responses_key_file = None
            responses_model = None
        if trigger_id is None and responses_key_file is None:
            raise ConfigurationError("no_central_model_channel")
        endpoint = sync.get("endpoint")
        token_file = sync.get("token_file")
        if endpoint is not None and (not isinstance(endpoint, str) or not endpoint.startswith(("http://", "https://"))):
            raise ConfigurationError("invalid_openclaw_sync_endpoint")
        if endpoint is not None:
            if not isinstance(token_file, str) or not Path(token_file).is_absolute():
                raise ConfigurationError("invalid_openclaw_sync_token_file")
            sync_token_file = Path(token_file)
        else:
            sync_token_file = None
        git_enabled = git_poll.get("enabled")
        if not isinstance(git_enabled, bool):
            raise ConfigurationError("invalid_git_poll_enabled")
        git_bare_repo_raw = git_poll.get("bare_repo")
        git_prefixes_raw = git_poll.get("ref_prefixes")
        if git_enabled:
            if not isinstance(git_bare_repo_raw, str) or not Path(git_bare_repo_raw).is_absolute():
                raise ConfigurationError("invalid_git_poll_bare_repo")
            if (
                not isinstance(git_prefixes_raw, list)
                or not git_prefixes_raw
                or any(not isinstance(item, str) or not item.startswith("refs/") for item in git_prefixes_raw)
            ):
                raise ConfigurationError("invalid_git_poll_ref_prefixes")
            git_bare_repo = Path(git_bare_repo_raw)
            git_prefixes = tuple(git_prefixes_raw)
        else:
            git_bare_repo = None
            git_prefixes = ()
        return cls(
            host=host,
            port=port,
            project_key=project_key,
            database_path=path_value("database_path"),
            event_inbox_dir=path_value("event_inbox_dir"),
            decision_inbox_dir=path_value("decision_inbox_dir"),
            dispatch_outbox_dir=path_value("dispatch_outbox_dir"),
            openclaw_sync_dir=path_value("openclaw_sync_dir"),
            command_queue_dir=path_value("command_queue_dir"),
            inbound_token_file=path_value("inbound_token_file"),
            callback_token_file=path_value("callback_token_file"),
            contract_commit=contract,
            workspace_trigger_id=trigger_id,
            workspace_token_file=workspace_token_file,
            responses_api_key_file=responses_key_file,
            responses_model=responses_model,
            poll_interval_seconds=integer(polling, "interval_seconds", 1, 300),
            summary_interval_seconds=integer(polling, "summary_interval_seconds", 10, 86400),
            callback_timeout_seconds=integer(loop, "callback_timeout_seconds", 30, 86400),
            lease_seconds=integer(loop, "lease_seconds", 10, 3600),
            retry_base_seconds=integer(loop, "retry_base_seconds", 1, 3600),
            max_retries=integer(loop, "max_retries", 0, 20),
            max_same_blocker_retries=integer(loop, "max_same_blocker_retries", 0, 20),
            max_hops_per_gate=integer(loop, "max_hops_per_gate", 1, 100),
            openclaw_sync_endpoint=endpoint,
            openclaw_sync_token_file=sync_token_file,
            git_poll_enabled=git_enabled,
            git_poll_bare_repo=git_bare_repo,
            git_poll_ref_prefixes=git_prefixes,
            git_poll_interval_seconds=integer(git_poll, "interval_seconds", 5, 3600),
        )
    def _secret(self, path: Path) -> str:
        owner_uid = int(os.environ.get("AIOS_ORCHESTRATOR_SECRET_OWNER_UID", "0"))
        info = verify_secure_file(path, expected_uid=owner_uid)
        if stat.S_IMODE(info.st_mode) & ~0o640:
            raise ConfigurationError(f"permissive_secret_mode:{path}")
        data = read_secure_bytes(path, expected_uid=owner_uid, max_bytes=64_000)
        if not data or b"\n" in data or b"\r" in data:
            raise ConfigurationError(f"invalid_secret_format:{path}")
        try:
            return data.decode("utf-8")
        except UnicodeDecodeError as exc:
            raise ConfigurationError(f"secret_not_utf8:{path}") from exc
    def inbound_token(self) -> str:
        return self._secret(self.inbound_token_file)
    def callback_token(self) -> str:
        return self._secret(self.callback_token_file)
    def workspace_token(self) -> str | None:
        return self._secret(self.workspace_token_file) if self.workspace_token_file else None
    def responses_api_key(self) -> str | None:
        return self._secret(self.responses_api_key_file) if self.responses_api_key_file else None
    def openclaw_sync_token(self) -> str | None:
        return self._secret(self.openclaw_sync_token_file) if self.openclaw_sync_token_file else None
@@ -1,122 +0,0 @@
 from __future__ import annotations
 import json
 from pathlib import Path
 import re
 import os
 import stat
 from typing import Any
 from .errors import DispatchError
 from .securefs import (
    atomic_write_bytes,
    atomic_write_bytes_shared,
    ensure_secure_directory,
    ensure_shared_directory,
    read_secure_bytes,
 )
 SAFE_ID = re.compile(r"^[A-Za-z0-9][A-Za-z0-9._:-]{0,127}$")
 class DispatchSink:
    """Reliable file-based dispatch adapter for existing OpenClaw/Codex watchers."""
    def __init__(self, outbox_dir: Path, command_queue_dir: Path, contract_commit: str):
        self.outbox_dir = Path(outbox_dir)
        self.command_queue_dir = Path(command_queue_dir)
        self.contract_commit = contract_commit
        ensure_secure_directory(self.outbox_dir)
        ensure_shared_directory(self.command_queue_dir)
    def dispatch(self, record: dict[str, Any]) -> tuple[Path, Path]:
        work_unit_id = record.get("work_unit_id")
        event_id = record.get("event_id")
        agent = record.get("assigned_agent")
        payload_json = record.get("payload_json")
        if (
            not isinstance(work_unit_id, str)
            or not SAFE_ID.fullmatch(work_unit_id)
            or not isinstance(event_id, str)
            or not SAFE_ID.fullmatch(event_id)
            or agent not in {"openclaw", "codex"}
            or not isinstance(payload_json, str)
        ):
            raise DispatchError("invalid_dispatch_record")
        try:
            work_unit = json.loads(payload_json)
        except json.JSONDecodeError as exc:
            raise DispatchError("invalid_dispatch_payload") from exc
        project_key = record.get("project_key")
        state_version = record.get("state_version")
        envelope = {
            "schema_version": 1,
            "contract_commit": self.contract_commit,
            "event_id": event_id,
            "assigned_agent": agent,
            "work_unit": work_unit,
            "callback": {
                "event_endpoint": "/v1/orchestrator/events",
                "start_event_type": "work_unit_started",
                "end_event_type": "work_unit_completed",
            },
        }
        outbox_path = self.outbox_dir / "dispatched" / f"{work_unit_id}.json"
        if agent == "codex":
            if (
                not isinstance(project_key, str)
                or not SAFE_ID.fullmatch(project_key)
                or not isinstance(state_version, int)
                or state_version < 0
            ):
                raise DispatchError("invalid_dispatch_context")
            required_checks = work_unit.get("required_checks", [])
            execution_kind = (
                "acceptance_probe"
                if isinstance(required_checks, list) and "acceptance_probe" in required_checks
                else "codex"
            )
            queue_value = {
                "schema_version": 1,
                "command_id": work_unit_id,
                "project_key": project_key,
                "work_unit_id": work_unit_id,
                "contract_commit": self.contract_commit,
                "state_version": state_version,
                "execution": {
                    "kind": execution_kind,
                    "input": work_unit.get("instruction", ""),
                },
            }
            # The Codex watcher polls the configured active directory itself.
            queue_path = self.command_queue_dir / f"{work_unit_id}.json"
        else:
            queue_value = envelope
            queue_path = self.command_queue_dir / agent / "pending" / f"{work_unit_id}.json"
        ensure_secure_directory(outbox_path.parent)
        ensure_shared_directory(queue_path.parent)
        # Idempotent immutable payload: existing identical file is accepted.
        self._write_idempotent_json(outbox_path, envelope, shared=False)
        self._write_idempotent_json(queue_path, queue_value, shared=True)
        return outbox_path, queue_path
    @staticmethod
    def _write_idempotent_json(path: Path, value: dict[str, Any], *, shared: bool) -> None:
        payload = (json.dumps(value, indent=2, sort_keys=True) + "\n").encode()
        try:
            st = os.lstat(path)
        except FileNotFoundError:
            if shared:
                atomic_write_bytes_shared(path, payload, mode=0o640)
            else:
                atomic_write_bytes(path, payload, mode=0o600)
            return
        if not stat.S_ISREG(st.st_mode):
            raise DispatchError(f"dispatch_target_not_regular:{path.name}")
        existing = read_secure_bytes(path, expected_uid=None, max_bytes=1_048_576)
        if existing != payload:
            raise DispatchError(f"dispatch_idempotency_conflict:{path.name}")
@@ -1,37 +0,0 @@
 class OrchestratorError(Exception):
    """Base orchestration error."""
 class ConfigurationError(OrchestratorError):
    """Unsafe or invalid configuration."""
 class ValidationError(OrchestratorError):
    """Invalid event or decision."""
 class AuthenticationError(OrchestratorError):
    """Authentication failed."""
 class ConflictError(OrchestratorError):
    """Idempotency, lease, or optimistic concurrency conflict."""
 class TriggerError(OrchestratorError):
    """Central model trigger failed."""
    def __init__(
        self,
        message: str,
        *,
        http_status: int | None = None,
        category: str | None = None,
    ) -> None:
        super().__init__(message)
        self.http_status = http_status
        self.category = category
 class DispatchError(OrchestratorError):
    """Dispatch or synchronization output failed."""
@@ -1,346 +0,0 @@
 from __future__ import annotations
 from dataclasses import dataclass, field
 import hashlib
 import json
 import re
 from typing import Any
 from .errors import ValidationError
 OID_RE = re.compile(r"^[0-9a-f]{40}$")
 ID_RE = re.compile(r"^[A-Za-z0-9][A-Za-z0-9._:-]{0,127}$")
 AGENTS = {"openclaw", "codex", "gitea", "system", "workspace-agent"}
 EVENT_STATUSES = {"READY", "RUNNING", "PASS", "BLOCKED", "FAIL", "TIMEOUT", "NOT_EXECUTED"}
 VERDICTS = {"DISPATCH", "BLOCKED", "COMPLETE", "NOOP"}
 EXECUTORS = {"openclaw", "codex"}
 def req_str(value: dict[str, Any], key: str, pattern: re.Pattern[str] | None = None) -> str:
    item = value.get(key)
    if not isinstance(item, str) or not item:
        raise ValidationError(f"invalid_{key}")
    if pattern is not None and not pattern.fullmatch(item):
        raise ValidationError(f"invalid_{key}")
    return item
 def opt_oid(item: Any, key: str) -> str | None:
    if item is None:
        return None
    if not isinstance(item, str) or not OID_RE.fullmatch(item):
        raise ValidationError(f"invalid_{key}")
    return item
 def string_list(value: dict[str, Any], key: str, limit: int) -> tuple[str, ...]:
    raw = value.get(key)
    if not isinstance(raw, list) or len(raw) > limit:
        raise ValidationError(f"invalid_{key}")
    result: list[str] = []
    for item in raw:
        if not isinstance(item, str) or not item or len(item) > 1024:
            raise ValidationError(f"invalid_{key}")
        result.append(item)
    return tuple(result)
@dataclass(frozen=True)
 class Event:
    schema_version: int
    project_key: str
    contract_commit: str
    event_id: str
    source_agent: str
    event_type: str
    work_unit_id: str
    status: str
    state_version: int
    implementation_commit: str | None = None
    result_commit: str | None = None
    blocker: str | None = None
    evidence_refs: tuple[str, ...] = ()
    payload: dict[str, Any] = field(default_factory=dict)
    @classmethod
    def from_dict(cls, value: dict[str, Any]) -> "Event":
        allowed = {
            "schema_version", "project_key", "contract_commit", "event_id",
            "source_agent", "event_type", "work_unit_id", "status",
            "state_version", "implementation_commit", "result_commit",
            "blocker", "evidence_refs", "payload",
        }
        if not isinstance(value, dict) or set(value) - allowed:
            raise ValidationError("invalid_event_fields")
        if value.get("schema_version") != 1:
            raise ValidationError("unsupported_event_schema")
        source = req_str(value, "source_agent")
        status = req_str(value, "status")
        if source not in AGENTS:
            raise ValidationError("invalid_source_agent")
        if status not in EVENT_STATUSES:
            raise ValidationError("invalid_event_status")
        version = value.get("state_version")
        if not isinstance(version, int) or version < 0:
            raise ValidationError("invalid_state_version")
        blocker = value.get("blocker")
        if blocker is not None and (not isinstance(blocker, str) or len(blocker) > 512):
            raise ValidationError("invalid_blocker")
        payload = value.get("payload", {})
        if not isinstance(payload, dict):
            raise ValidationError("invalid_payload")
        if len(json.dumps(payload, sort_keys=True).encode()) > 262_144:
            raise ValidationError("payload_too_large")
        return cls(
            schema_version=1,
            project_key=req_str(value, "project_key", ID_RE),
            contract_commit=req_str(value, "contract_commit", OID_RE),
            event_id=req_str(value, "event_id", ID_RE),
            source_agent=source,
            event_type=req_str(value, "event_type", ID_RE),
            work_unit_id=req_str(value, "work_unit_id", ID_RE),
            status=status,
            state_version=version,
            implementation_commit=opt_oid(value.get("implementation_commit"), "implementation_commit"),
            result_commit=opt_oid(value.get("result_commit"), "result_commit"),
            blocker=blocker,
            evidence_refs=string_list(value, "evidence_refs", 64),
            payload=payload,
        )
    def to_dict(self) -> dict[str, Any]:
        return {
            "schema_version": 1,
            "project_key": self.project_key,
            "contract_commit": self.contract_commit,
            "event_id": self.event_id,
            "source_agent": self.source_agent,
            "event_type": self.event_type,
            "work_unit_id": self.work_unit_id,
            "status": self.status,
            "state_version": self.state_version,
            "implementation_commit": self.implementation_commit,
            "result_commit": self.result_commit,
            "blocker": self.blocker,
            "evidence_refs": list(self.evidence_refs),
            "payload": self.payload,
        }
    def canonical_json(self) -> str:
        return json.dumps(self.to_dict(), separators=(",", ":"), sort_keys=True)
    def digest(self) -> str:
        return hashlib.sha256(self.canonical_json().encode()).hexdigest()
@dataclass(frozen=True)
 class WorkUnit:
    id: str
    track_id: str
    gate_id: str
    primary_agent: str
    fallback_agent: str
    clean_base_commit: str
    allowed_paths: tuple[str, ...]
    required_checks: tuple[str, ...]
    forbidden_operations: tuple[str, ...]
    instruction: str
    @classmethod
    def from_dict(cls, value: dict[str, Any]) -> "WorkUnit":
        expected = {
            "id", "track_id", "gate_id", "primary_agent", "fallback_agent",
            "clean_base_commit", "allowed_paths", "required_checks",
            "forbidden_operations", "instruction",
        }
        if not isinstance(value, dict) or set(value) != expected:
            raise ValidationError("invalid_work_unit_fields")
        primary = req_str(value, "primary_agent")
        fallback = req_str(value, "fallback_agent")
        if primary not in EXECUTORS or fallback not in EXECUTORS or primary == fallback:
            raise ValidationError("invalid_agent_assignment")
        instruction = req_str(value, "instruction")
        if len(instruction.encode()) > 65_536:
            raise ValidationError("instruction_too_large")
        return cls(
            id=req_str(value, "id", ID_RE),
            track_id=req_str(value, "track_id", ID_RE),
            gate_id=req_str(value, "gate_id", ID_RE),
            primary_agent=primary,
            fallback_agent=fallback,
            clean_base_commit=req_str(value, "clean_base_commit", OID_RE),
            allowed_paths=string_list(value, "allowed_paths", 128),
            required_checks=string_list(value, "required_checks", 128),
            forbidden_operations=string_list(value, "forbidden_operations", 64),
            instruction=instruction,
        )
    def to_dict(self) -> dict[str, Any]:
        return {
            "id": self.id,
            "track_id": self.track_id,
            "gate_id": self.gate_id,
            "primary_agent": self.primary_agent,
            "fallback_agent": self.fallback_agent,
            "clean_base_commit": self.clean_base_commit,
            "allowed_paths": list(self.allowed_paths),
            "required_checks": list(self.required_checks),
            "forbidden_operations": list(self.forbidden_operations),
            "instruction": self.instruction,
        }
@dataclass(frozen=True)
 class Decision:
    schema_version: int
    project_key: str
    contract_commit: str
    event_id: str
    expected_state_version: int
    verdict: str
    state_updates: dict[str, str]
    next_work_units: tuple[WorkUnit, ...]
    requires_human_input: bool
    stop_reason: str | None
    @classmethod
    def from_dict(cls, value: dict[str, Any]) -> "Decision":
        expected = {
            "schema_version", "project_key", "contract_commit", "event_id",
            "expected_state_version", "verdict", "state_updates",
            "next_work_units", "requires_human_input", "stop_reason",
        }
        if not isinstance(value, dict) or set(value) != expected:
            raise ValidationError("invalid_decision_fields")
        if value.get("schema_version") != 1:
            raise ValidationError("unsupported_decision_schema")
        version = value.get("expected_state_version")
        if not isinstance(version, int) or version < 0:
            raise ValidationError("invalid_expected_state_version")
        verdict = req_str(value, "verdict")
        if verdict not in VERDICTS:
            raise ValidationError("invalid_verdict")
        updates = value.get("state_updates")
        normalized: dict[str, str] = {}
        if isinstance(updates, dict):
            update_items = [{"key": key, "value": item} for key, item in updates.items()]
        elif isinstance(updates, list):
            update_items = updates
        else:
            raise ValidationError("invalid_state_updates")
        if len(update_items) > 128:
            raise ValidationError("invalid_state_updates")
        for update in update_items:
            if not isinstance(update, dict) or set(update) != {"key", "value"}:
                raise ValidationError("invalid_state_update")
            key = update.get("key")
            item = update.get("value")
            if not isinstance(key, str) or not ID_RE.fullmatch(key):
                raise ValidationError("invalid_state_update_key")
            if not isinstance(item, str) or not item or len(item) > 256:
                raise ValidationError("invalid_state_update_value")
            if key in normalized:
                raise ValidationError("duplicate_state_update_key")
            normalized[key] = item
        raw_units = value.get("next_work_units")
        if not isinstance(raw_units, list) or len(raw_units) > 32:
            raise ValidationError("invalid_next_work_units")
        units = tuple(WorkUnit.from_dict(item) for item in raw_units)
        human = value.get("requires_human_input")
        if not isinstance(human, bool):
            raise ValidationError("invalid_requires_human_input")
        stop = value.get("stop_reason")
        if stop is not None and (not isinstance(stop, str) or len(stop) > 512):
            raise ValidationError("invalid_stop_reason")
        if verdict == "DISPATCH" and not units:
            raise ValidationError("dispatch_without_work_units")
        if verdict != "DISPATCH" and units:
            raise ValidationError("work_units_without_dispatch")
        return cls(
            schema_version=1,
            project_key=req_str(value, "project_key", ID_RE),
            contract_commit=req_str(value, "contract_commit", OID_RE),
            event_id=req_str(value, "event_id", ID_RE),
            expected_state_version=version,
            verdict=verdict,
            state_updates=normalized,
            next_work_units=units,
            requires_human_input=human,
            stop_reason=stop,
        )
    def to_dict(self) -> dict[str, Any]:
        return {
            "schema_version": 1,
            "project_key": self.project_key,
            "contract_commit": self.contract_commit,
            "event_id": self.event_id,
            "expected_state_version": self.expected_state_version,
            "verdict": self.verdict,
            "state_updates": [
                {"key": key, "value": value}
                for key, value in sorted(self.state_updates.items())
            ],
            "next_work_units": [u.to_dict() for u in self.next_work_units],
            "requires_human_input": self.requires_human_input,
            "stop_reason": self.stop_reason,
        }
 DECISION_SCHEMA: dict[str, Any] = {
    "type": "object",
    "additionalProperties": False,
    "required": [
        "schema_version", "project_key", "contract_commit", "event_id",
        "expected_state_version", "verdict", "state_updates",
        "next_work_units", "requires_human_input", "stop_reason",
    ],
    "properties": {
        "schema_version": {"type": "integer", "const": 1},
        "project_key": {"type": "string"},
        "contract_commit": {"type": "string", "pattern": "^[0-9a-f]{40}$"},
        "event_id": {"type": "string"},
        "expected_state_version": {"type": "integer", "minimum": 0},
        "verdict": {"type": "string", "enum": sorted(VERDICTS)},
        "state_updates": {
            "type": "array",
            "items": {
                "type": "object",
                "additionalProperties": False,
                "required": ["key", "value"],
                "properties": {
                    "key": {"type": "string"},
                    "value": {"type": "string"}
                }
            }
        },
        "next_work_units": {
            "type": "array",
            "maxItems": 32,
            "items": {
                "type": "object",
                "additionalProperties": False,
                "required": [
                    "id", "track_id", "gate_id", "primary_agent",
                    "fallback_agent", "clean_base_commit", "allowed_paths",
                    "required_checks", "forbidden_operations", "instruction",
                ],
                "properties": {
                    "id": {"type": "string"},
                    "track_id": {"type": "string"},
                    "gate_id": {"type": "string"},
                    "primary_agent": {"type": "string", "enum": ["openclaw", "codex"]},
                    "fallback_agent": {"type": "string", "enum": ["openclaw", "codex"]},
                    "clean_base_commit": {"type": "string", "pattern": "^[0-9a-f]{40}$"},
                    "allowed_paths": {"type": "array", "items": {"type": "string"}},
                    "required_checks": {"type": "array", "items": {"type": "string"}},
                    "forbidden_operations": {"type": "array", "items": {"type": "string"}},
                    "instruction": {"type": "string"},
                },
            },
        },
        "requires_human_input": {"type": "boolean"},
        "stop_reason": {"type": ["string", "null"]},
    },
 }
--- a/Show More
+++ b/Show More
Author	SHA1	Message	Date
Codex	81572b8231	security(openclaw): record restored Gitea credential without secret value	2026-07-20 15:37:29 +08:00
Codex	8c3d80fe53	task(openclaw): 下发执行器启动、Secret 引用与 Gitea 同步在线指令	2026-07-20 14:34:13 +08:00
Codex	99bcc03b40	task(openclaw): 初始化无人值守任务状态	2026-07-20 12:07:46 +08:00
Codex	c6e7edcf54	task(openclaw): 写入预发布无人值守任务载荷	2026-07-20 12:07:36 +08:00
Codex	c464560a0f	task(openclaw): launch AIOS V2R1 pre-production deployment	2026-07-20 11:49:09 +08:00
Codex	f23b694ad3	tasks(aios-v2): add Codex production supervisor task	2026-07-20 11:46:43 +08:00
Codex	f97dea55ac	tasks(aios-v2): add OpenClaw production deployment task	2026-07-20 11:46:39 +08:00
Codex	efd08d7771	docs(aios-v2): add ADR-003 GitOps-only flow	2026-07-20 11:46:31 +08:00
Codex	ff6cb2e217	docs(aios-v2): add ADR-005 documentation sync gate	2026-07-20 11:45:16 +08:00
Codex	d774620fea	docs(aios-v2): add ADR-002 runtime decoupling decision	2026-07-20 11:45:12 +08:00
Codex	b26ff010fc	docs(aios-v2): add ADR-001 control plane decision	2026-07-20 11:45:09 +08:00
Codex	4b49f2ba4b	docs(aios-v2): add ADR index	2026-07-20 11:45:04 +08:00
Codex	b4b8fe8828	docs(aios-v2): add traceability matrix	2026-07-20 11:43:46 +08:00
Codex	4e8ab051a0	docs(aios-v2): add module contracts	2026-07-20 11:41:39 +08:00
Codex	f61e35b218	docs(aios-v2): add migration and delivery strategy	2026-07-20 11:34:33 +08:00
Codex	b2f90f24a3	docs(aios-v2): add target architecture	2026-07-20 11:32:16 +08:00
Codex	6721eb4752	docs(aios-v2): add user requirements baseline	2026-07-20 11:29:07 +08:00
Codex	a9e4fd1e40	docs(aios-v2): establish architecture source of truth	2026-07-20 11:24:12 +08:00
Codex	621750976a	deploy(candidate): add guarded release rollback	2026-07-20 11:22:04 +08:00
Codex	5359e88806	feat(candidate): expose sanitized GitOps status API	2026-07-20 11:19:37 +08:00
Codex	76636931ba	feat(candidate): add sanitized GitOps status provider	2026-07-20 11:17:18 +08:00
Codex	180c1fc66e	deploy(candidate): add end-to-end healthcheck	2026-07-20 11:14:53 +08:00
Codex	c0ff052dab	deploy(candidate): add Docker Compose runtime	2026-07-20 11:12:42 +08:00
Codex	94cffb6ba3	deploy(candidate): add hardened container image	2026-07-20 11:10:36 +08:00
Codex	9ad5628e69	deploy(candidate): add environment configuration template	2026-07-20 11:08:37 +08:00
Codex	6d397a7894	deploy(candidate): add hardened systemd service	2026-07-20 11:07:06 +08:00
Codex	165fea6ba9	fix(candidate): prevent Hermes completion heartbeat revision race	2026-07-20 11:05:45 +08:00
Codex	d079e2087c	test(candidate): add end-to-end runtime and persistence tests	2026-07-20 11:01:07 +08:00
Codex	e3b36ebdcc	feat(candidate): add production application entrypoint	2026-07-20 10:57:42 +08:00
Codex	5e791b11c0	feat(candidate): add tenant-safe web console	2026-07-20 10:54:43 +08:00
Codex	50be65b76e	feat(candidate): add durable SQLite control-plane persistence	2026-07-20 10:50:31 +08:00
Codex	7bb86e30bf	feat(candidate): export orchestration and execution tracking	2026-07-20 10:42:49 +08:00
Codex	ea56b45b7e	feat(candidate): add persistent agent execution tracking	2026-07-20 10:40:22 +08:00
Codex	f53682da7a	feat(candidate): add unified AIOS execution orchestrator	2026-07-20 10:38:06 +08:00
Codex	71884ac088	feat(candidate): add Hermes worker loop and lease heartbeat	2026-07-20 10:35:55 +08:00
Codex	969c278cb0	feat(candidate): implement Hermes queue and lease adapter	2026-07-20 10:34:25 +08:00
Codex	6fe2a1e036	feat(candidate): expose Hermes runtime package	2026-07-20 10:31:07 +08:00
Codex	9a60e0c68e	feat(candidate): add Hermes runtime contracts	2026-07-20 10:27:47 +08:00
Codex	46f553ca11	feat(candidate): expose Hermes runtime boundary package	2026-07-20 10:17:02 +08:00
Codex	4f0a0c2184	feat(candidate): add OpenClaw audit callback sink	2026-07-20 10:07:54 +08:00
Codex	c107b193d9	feat(candidate): add OpenClaw runtime health policy	2026-07-20 10:06:36 +08:00
Codex	2db7ced7a4	feat(candidate): add OpenClaw runtime transports	2026-07-20 10:01:59 +08:00
Codex	6361d9a334	feat(candidate): implement OpenClaw execution adapter	2026-07-20 09:59:53 +08:00
Codex	2cf54f04ef	feat(candidate): expose OpenClaw adapter package	2026-07-20 09:56:06 +08:00
Codex	f840ce5c94	feat(integration): add OpenClaw command contract	2026-07-20 09:49:15 +08:00
Codex	35e7b55490	integration(aios-v2): create CTO production candidate branch	2026-07-20 09:47:01 +08:00
Codex	1f8d1f23b3	feat(aios-v2): add tenant-aware HTTP control plane gateway	2026-07-20 02:24:19 +08:00
Codex	e93011791e	feat(aios-v2): add tenant-isolated control plane domain	2026-07-20 02:19:16 +08:00
Codex	4460564a48	feat(aios-v2): initialize control plane package	2026-07-20 01:11:11 +08:00
Codex	4d6658136e	docs(task): add AIOS control plane gateway task contract	2026-07-20 01:06:11 +08:00
Codex	5767e665c2	task(codex): implement AIOS V2 control plane gateway v1	2026-07-20 01:06:04 +08:00
		`@@ -1,3 +0,0 @@`
			`"""Reliable central orchestration gateway for AIOS."""`

			`__version__ = "0.2.0"`